EP0827342B1 - Prädiktive Bildkodierung - Google Patents
Prädiktive Bildkodierung Download PDFInfo
- Publication number
- EP0827342B1 EP0827342B1 EP97907276A EP97907276A EP0827342B1 EP 0827342 B1 EP0827342 B1 EP 0827342B1 EP 97907276 A EP97907276 A EP 97907276A EP 97907276 A EP97907276 A EP 97907276A EP 0827342 B1 EP0827342 B1 EP 0827342B1
- Authority
- EP
- European Patent Office
- Prior art keywords
- encoding
- decoding
- mode
- picture element
- prediction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Lifetime
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
- H04N19/593—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving spatial prediction techniques
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/41—Bandwidth or redundancy reduction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/30—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using hierarchical techniques, e.g. scalability
Definitions
- the present invention relates to an encoding apparatus, a decoding apparatus and their methods for encoding and decoding picture information generated by and used in a facsimile machine, a scanner, and a computer. More specifically, the present invention relates to an encoding apparatus and a decoding apparatus having two types of encoding and decoding systems for efficiently encoding and decoding picture information by switching between the two types of encoding and decoding systems. Further, the present invention relates to an encoding method and a decoding method for encoding and decoding picture information efficiently.. The present invention also relates to a picture processing apparatus having an encoding apparatus and a decoding apparatus according to the present invention. The present invention also relates to a picture processing apparatus for implementing an encoding method and a decoding method according to the present invention.
- Fig. 66 is a block diagram showing a conventional encoding apparatus.
- reference numeral 901 indicates a picture element memory for receiving, storing, and outputting the value of a picture element to be encoded (which will be referred to as an encoding picture element, or simply as a picture element) and for outputting the value of at least one encoded picture element already stored in the picture element memory and adjacent to the encoding picture element as the value of a reference picture element.
- Reference numeral 907 indicates a predictor for calculating the prediction value for the encoding picture element by referring to the value of the at least one reference picture element.
- Reference numeral 931 indicates a prediction error calculator for determining the prediction error by subtracting the prediction value calculated by the predictor 907 from the value of the encoding picture element.
- Reference numeral 908 indicates an encoder for encoding the prediction error between the value of the encoding picture element and the prediction value calculated by the predictor 907, and for outputting codewords.
- Reference numeral 910 indicates a code buffer for receiving the codewords supplied from the encoder 908 and for outputting a sequence of the codewords as a code in order of the received codewords.
- the predictor 907 calculates the prediction value from the value of the at least one reference picture element.
- the calculation method may be implemented in accordance with a predetermined prediction function or by referring to a reference table.
- the encoder 908 encodes the prediction error (-255 ⁇ +255, inclusive of zero, in the case of one picture element being represented by eight bits) which has been obtained by subtracting the calculated prediction value from the value of an encoding picture element by using a predetermined codeword table.
- one codeword is allotted to a binary symbol sequence composed of one binary symbol or a plurality of binary symbols.
- the term "encoding” is used in this specification to mean an operation for determining and allotting a codeword to a sequence of a certain number (which will be hereinafter referred to as the code order) of binary "0" symbols (More Probable Symbols abbreviated to MPSs) or binary "1" symbols (Less Probable Symbol abbreviated to LPSs) occurred, and for outputting the codeword therefor.
- the number of MPSs consecutively occurred is counted by an MPS counter (not shown) inside (or outside) the encoder.
- the counted value of MPSs is stored in an MPS memory (not shown), and the state numbers of binary symbol sequences (to be described hereafter) are stored in a state-number memory (not shown).
- the code order may be an integer greater than zero. However, it is assumed herein that the code order is restricted to 2 n (the n-th power of 2).
- the number of MPSs consecutively occurred has become equal to the code order 2 n , one-bit codeword "0" is allotted to the MPSs.
- the number of the MPSs consecutively occurred after outputting the latest codeword before occurring the LPS is expressed in terms of n-bit binary symbols, and, in.order to differentiate from the sequence of the MPSs to which the codeword "0" is allotted, the codeword "1" is added to the beginning of the n-bit binary symbols. Accordingly, a codeword of (n+1) bits is allotted to the sequence of MPSs plus the LPS differentiating from the sequence of MPSs to which a codeword of "0" is allotted.
- the unit of a binary symbol sequence to which a codeword is allotted is.referred to as a message.
- the MPS counter is reset.
- a sequence of codewords output in this way constitutes a code.
- the code is supplied to the decoder and divided into individual codewords. Then, a binary symbol sequence is recreated by the decoder, and picture elements are reproduced. In this way, decoding is implemented.
- the code order is changed so as to represent the appropriate code length in accordance with the occurrence probability of one of binary symbols estimated from past data on binary symbol sequences. For this reason, a further excellent encoding efficiency can be obtained.
- the binary symbol sequence belongs to one of the sixteen states shown in Fig. 68 .
- the code order is determined according to the state to which each binary symbol sequence belongs. It is assumed herein that the initial value of the state number for the encoder or the decoder is set to 0. It is also assumed herein that the MPS counter of the encoder.or the decoder is reset at the beginning of the encoding process or the decoding process. During the encoding process or the decoding process, the encoder or the decoder implements state transition when a codeword has been determined.
- the state number of the sequence is increased by one.
- the state number of the sequence is decreased by one.
- the encoder or the decoder does not implement state transition, and the state number remains uncharged.
- a second example of the method of determining the code order there is shown a method in which the numbers of binary symbols "0" and binary symbols "1" which have occurred in a binary symbol sequence, respectively indicated by N(0) and N(1), are counted on both the transmitting side and the receiving side within the same range (such as, for example, within one line) so as to calculate the code order of the binary symbol sequence, based on the result of the count.
- the method of determining the code order for example, is disclosed in Japanese Unexamined Patent Publication No. SHO59-27501 (which corresponds to United States Patent No. 4191974 ).
- the calculation method is expressed by the relation of 2 n+1 N(1) > N(0) ⁇ 2 n N(1). In this case, however, the code order 2" which varies with state transition of a binary symbol sequence is not less than a predetermined minimum values, nor more than a predetermined maximum value.
- Fig. 67 It is known that the encoding method shown in Fig. 67 has the following characteristics. Let us assume that a binary information source whose probability of binary symbol "0" and whose probability of binary symbol "1" are p, 1-p (p ⁇ 1/2 ) respectively are encoded in accordance with the encoding method shown in Fig. 67 .
- the order n rendering a maximum code length in each code order minimum fulfills the following expression: 2 n / 2 n + 1 ⁇ p ⁇ 2 n + 1 / 2 n + 1 + 1
- an encoding apparatus or a decoding apparatus wherein two encoding or decoding modes such as the mode A and the mode B are provided, for example, and encoding or decoding is implemented by switching between the mode A and the mode B according to the decision whether or not a predetermined condition for the value(s) of reference picture element(s) is satisfied. Basically, if the value(s) of reference picture element(s) satisfies a predetermined condition, encoding or decoding is implemented in the mode A. On the other hand, if the value(s) of reference picture element(s) does not satisfy a predetermined condition, encoding or decoding is implemented in the mode B.
- Mode switching may be accomplished; for example, in accordance with the method described in " The National Assembly 1016 of the Institute of Electronics and Communication Engineers of Japan held in 1977 " as the "run length encoding process according to the encoding start patterns” .
- the mode A is switched into the mode B. Then, the picture element X and the subsequent picture elements are encoded or decoded continuously in the mode B.
- the mode is switched to the mode A and the subsequent encoding or decoding picture element is encoded or decoded in the mode A.
- Fig. 70 shows a configuration of the picture compression circuit and the picture expansion circuit.
- the picture compression circuit includes a process for implementing a lossless compression and another process for implementing a lossy compression.
- a lossy picture compression means a compressing process where a compressibility of the picture is increased, though a quality of reproduced picture (reproducibility) is decreased.
- a lossless picture compression means a compressing process where the quality of reproduced picture (reproducibility) is not decreased, though a compressibility of the picture is less increased than the above lossy picture compression.
- a DCT (Discrete Cosine Transform) circuit 951 performs two-dimensional DCT operation on an input picture to divide the picture into two-dimensional spatial frequency components.
- a quantization circuit 952 quantizes a DCT coefficient.
- An entropy encoder 953 implements Huffman coding on the quantized DCT coefficient.
- the lossy picture compression is performed by the DCT circuit 951, the quantization circuit 952 and the entropy encoder 953.
- a predictor 954 predicts data of a certain picture element by using data of the previous picture element.
- An entropy encoder 955 implements Huffman coding of a differential between the picture element and the picture element predicted by the predictor 954. In this way, the lossless data compression is implemented by the predictor 954 and the entropy encoder 955.
- a switch SW1 selects one of the compressing processes: side "a" of the lossless compression; and side "b" of the lossy compression.
- the picture expansion circuit includes a process for implementing a lossless expansion and a process for implementing lossy expansion.
- An entropy decoder 956 and a decoder 957 decode the reversibly compressed data by an inverse operation of the entropy encoder 955 and the predictor 954.
- An entropy decoder 958, a dequantization circuit 959 and an inverse DCT circuit 960 decode the compressed data by an inverse operation of the DCT circuit 951, the quantization circuit 952 and the entropy encoder 953.
- a switch SW2 selects one of the expanding processes: side "a" of the lossless expansion; and side "b" of the lossy expansion.
- the conventional encoding apparatus which has been described as the related art 1 encodes a prediction error by referring to a predetermined codeword table.
- the statistical characteristic of the picture information displayed on a screen varies greatly depending on the part on the screen. In other words, it is known that it occurs that prediction for some part of the picture information displayed on the screen tends to be correct while other part of the picture information displayed on the screen often has great prediction errors.
- the encoding apparatus according to the first conventional related art implements encoding by referring to a single codeword table. Thus, it has created a problem in that an encoding efficiency cannot be enhanced.
- the encoding method which has been described as the related art 2 is a method of implementing encoding by referring to a plurality of codeword tables and dynamically changing the code order depending on the occurrence probability of the MPS. Consequently, if the statistical characteristic of picture information displayed varies greatly, a more excellent encoding efficiency will be provided with this encoding system than with the encoding apparatus which has been described as the first conventional related art.
- the encoding method according to related art 2 when at least one codeword is allotted to a prediction error for each encoding picture element, at least one-bit code amount is required for each picture element regardless of whether or not the prediction has proved to be correct (or no prediction error has been produced).
- Allotting a one-bit or more bits of codeword to a prediction error although the prediction probability exceeds 1/2 means that the actual code amount required is greater than the theoretical minimum code amount (entropy) for the prediction error. In other words, it means that an encoding efficiency is reduced.
- the picture compression circuit (encoding apparatus) and the picture expansion circuit (decoding apparatus) are configured as shown in Fig. 70 .
- the DCT circuit, the quantization circuit and the entropy encoder implement lossy picture compression and lossy picture expansion.
- the predictor and the entropy encoder implement lossless picture compression and lossless picture expansion.
- the conventional picture pick-up apparatus switches the lossless picture compression circuit and the lossy picture compression circuit according to the condition. In the picture pick-up apparatus, it is mostly required to increase the compressibility of the picture without decreasing the quality of the reproduced picture (reproducibility).
- picture information On handling multimedia information, picture information is transmitted, displayed, or stored together with other information such as audio information, or character information. Picture information occupies higher ratio than other information among such multimedia information, and these days it is required that the compressibility of picture is further increased.
- the present invention has been made to solve the above-mentioned problems. It is therefore an object of the present invention to provide an encoding apparatus and a decoding apparatus which can implement encoding and decoding picture information efficiently.
- An encoding apparatus may comprise:
- the first encoding section may comprise:
- the second encoding section may comprise:
- the first encoder may comprise a first probability estimator for receiving the result of the determination output from the determinator as a sequence of binary symbols and estimating an occurrence probability of one of the binary symbols, and a first codeword allotter for encoding the sequence of binary symbols
- the second encoder may comprise a first error-to-symbol converter for receiving the prediction error and converting the prediction error into a sequence of binary symbols
- a second probability estimator for receiving the sequence of binary symbols and estimating an occurrence probability of one of the binary symbols
- the third encoder may comprise a second error-to-symbol converter for receiving the prediction error and converting the prediction error into a sequence of binary symbols
- a third probability estimator for receiving the sequence of binary symbols and estimating an occurrence probability of one of the binary symbols
- a third codeword allotter for encoding the sequence of binary symbols.
- the encoding controller may have a codeword transmission order controller for changing an order for outputting codewords when a codeword has been-determined by at least one of the first encoder, the second encoder, and the third encoder and when a codeword has not been determined by the other encoders.
- At least one of the first error-to-symbol converter and the second error-to-symbol converter successively may generate comparison values starting with a value which would most likely occur as a prediction error to be input to one of the first error-to-symbol converter and the second error-to-symbol converter, may successively compare the generated values with the prediction error input to one of the first error-to-symbol converter and the second error-to-symbol converter one by one, and may generate and output a sequence of binary symbols based on a count of comparison time until one of the generated comparison values coincides with the prediction error.
- the mode determinator may select an encoding mode for an encoding picture element based on an encoding mode for an encoded picture element preceding the encoding picture element.
- At least one of the first codeword allotter, the second codeword allotter, and the third codeword allotter may change interpretation as to which one of the binary symbols is a more probable symbol, based on a change in probability estimation for one of the binary symbols respectively implemented by the first probability estimator for the first codeword allotter, the second probability estimator for the second codeword allotter, and the third probability estimator for the third codeword allotter.
- the first predictor of the first encoding section and the second predictor of the second encoding section may be combined into a predictor.
- At least two encoders of the first encoder, the second encoder, and the third encoder may be combined into an encoder.
- At least one of the first codeword allotter, the second codeword allotter, and the third codeword allotter may implement encoding for an enlarged information source of the binary symbols effected by selecting a code most suited to a state of the enlarged information source of the binary symbols assumed from the estimated occurrence probability of the more probable symbol from a Huffman code set prepared systematically for the enlarged information source of the binary symbols.
- a decoding apparatus may comprise:
- the first decoding section may comprise:
- the second decoding section may comprise:
- the first decoder may have a first symbol restoring device for receiving the codeword and decoding the codeword into a sequence of binary symbols and a first probability estimator for estimating an occurrence probability of one of the binary symbols, and wherein the first decoder outputs one of the binary symbols as a result of determination
- the second decoder may have a second symbol restoring device for receiving the codeword and decoding the codeword into a sequence of binary symbols, a second probability estimator for receiving the binary symbols and estimating an occurrence probability of one of the binary symbols, and a first symbol-to-error converter for receiving the sequence of binary symbols and converting the sequence of binary symbols into the prediction error
- the third decoder may have a third symbol restoring device for receiving the codeword and decoding the codeword into a sequence of binary symbols, a third probability estimator for receiving the binary symbols and estimating an occurrence probability of one of the binary symbols, and a second symbol-to-error converter for receiving the sequence of binary symbols and converting the sequence of binary symbols into the prediction error.
- the decoding controller may have a binary-symbol sequence using order controller for changing an order of using decoded binary symbols in a case where, before all sequences of binary symbols decoded by at least one of the first decoder, the second decoder, and the third decoder are used up, a sequence of binary symbols is output from the other one of the first decoder, the second decoder, and the third decoder.
- At least one of the first symbol-to-error converter and the second symbol-to-error converter may convert the input sequence of binary symbols into the prediction error based on a value and a number of the binary symbols input.
- the mode determinator may select a decoding mode for a decoding picture element based on a decoding mode for a decoded picture element preceding the decoding picture element.
- At least one of the first symbol restoring device, the second symbol restoring device, and the third symbol restoring device may change interpretation as to which one of the binary symbols is a more probable symbol based on a change in probability estimation for the binary symbols respectively implemented by the first probability estimator for the first symbol restoring device, the second probability estimator for the second symbol restoring device and the third probability estimator for the third symbol restoring device.
- the first predictor of the first decoding section and the second predictor of the second decoding section may be combined into a predictor.
- At least two decoders of the first decoder, the second decoder, and the third decoder may be combined into a decoder.
- At least one of the first symbol restoring device, the second symbol restoring device, and the third symbol restoring device may implement decoding for an enlarged information source of binary symbols effected by selecting a code most suited to a state of the enlarged information source of binary symbols assumed from the estimated occurrence probability of the more probable symbol from a Huffman code set prepared systematically for the enlarged information source of binary symbols.
- An encoding method according to the present invention may comprise:
- the first main encoding step may comprise:
- the second main encoding step may comprise:
- the first encoding step may comprise a first probability estimating step of receiving the result of the determination output by the determination step as a sequence of binary symbols and estimating an occurrence probability of one of the binary symbols and a first codeword allotting step of encoding the sequence of binary symbols
- the second encoding step may comprise a first error-to-symbol converting step of receiving the prediction error and converting the prediction error into a sequence of binary symbols
- the third encoding step may comprise a second error-to-symbol converting step of receiving the prediction error and converting the prediction error into a sequence of binary symbols, a third probability estimating step of receiving the sequence of binary symbols and estimating an occurrence probability of one of the binary symbols, and a third codeword allotting step of encoding the sequence of binary symbols.
- At least one of the first codeword allotting step, the second codeword allotting step, and the third codeword allotting step may include a step of changing interpretation as to which one of the binary symbols is a more probable symbol based on a change in probability estimation for the binary symbols respectively implemented by the first probability estimating step for the first codeword allotting step, the second probability estimating step for the second codeword allotting step, and the third probability estimating step for the third codeword allotting step.
- At least one of the first codeword allotting step, the second codeword allotting step, and the third codeword allotting step may implement encoding for an enlarged information source of the binary symbols effected by selecting a code most suited to a state of the enlarged information source of the binary symbols assumed from the estimated occurrence probability of the more probable symbol from a Huffman code set prepared systematically for the enlarged information source of the binary symbols.
- a decoding method according to the present invention may comprise:
- the first main decoding step may comprise:
- the second main decoding step may comprise:
- the first decoding step may comprise a first symbol restoring step for receiving the codeword and decoding the codeword into a sequence of binary symbols and a first probability estimating step of estimating an occurrence probability of one of the binary symbols, and a step of outputting one of the binary symbols as a result of the determination
- the second decoding step may comprise a second symbol restoring step of receiving the codeword and decoding the codeword into a sequence of binary symbols, a second probability estimating step of estimating an occurrence probability of one of the binary symbols, and a first symbol-to-error converting step of receiving the sequence of binary symbols and converting the sequence of binary symbols into the prediction error
- the third decoding step may comprise a third symbol restoring step of receiving the codeword and decoding the codeword into a sequence of binary symbols, a third probability estimating step of receiving the binary symbols and estimating an occurrence probability of one of the binary symbols, and a second symbol-to-error converting step of receiving the sequence of binary symbols and converting the sequence of binary symbols into the prediction error.
- At least one of the first symbol restoring step, the second symbol restoring step, and the third symbol restoring step may include a step of changing interpretation as to which one of the binary symbols is a more probable symbol based on a change in probability estimation for the binary symbols respectively implemented by the first probability estimating step for the first symbol restoring step, the second probability estimating step for the second symbol restoring step, and the third probability estimating step for the third symbol restoring step.
- At least one of the first symbol restoring step, the second symbol restoring step and the third symbol restoring step may implement decoding for an enlarged information source of binary symbols effected by selecting a code most suited to a state of the enlarged information source of binary symbols assumed from the estimated occurrence probability of the more probable symbol from a Huffman code set prepared systematically for the enlarged information source of binary symbols.
- the encoding apparatus may be provided in a semiconductor chip.
- the encoding apparatus may be provided in a circuit board.
- a picture encoding apparatus for receiving picture signals representing picture elements, encoding the picture elements represented by the picture signals into codes by using an encoding apparatus therein, and for outputting the.codes to a subsequent processing apparatus, the encoding apparatus may comprise:
- the picture processing apparatus may be an electronic computer.
- the picture processing apparatus may be a scanner.
- the picture processing apparatus may be a facsimile machine.
- the picture processing apparatus may be a display unit.
- the picture processing apparatus may be a storage device.
- the decoding apparatus may be provided in a semiconductor chip.
- the decoding apparatus may be provided in a circuit board.
- a picture processing apparatus for receiving a picture signal representing a code for a picture element, decoding the code into the value of the picture element by a decoding apparatus therein, and for outputting the picture element to a subsequent processing apparatus, the decoding apparatus may comprise:
- the picture processing apparatus may be an electronic computer.
- the picture processing apparatus may be a scanner.
- the picture processing apparatus may be a facsimile machine.
- the picture processing apparatus may be a printer.
- the picture processing apparatus may be a display unit.
- the picture processing apparatus may be a storage device.
- the encoding apparatus may comprise:
- the encoding section may comprise:
- a decoding apparatus may comprise:
- the decoding section may comprise:
- a picture processing apparatus may comprise:
- a picture processing apparatus may comprise:
- the picture compression circuit may comprise a plurality of the encoding apparatuses for inputting a luminance signal Y and color-difference signals U, V in parallel and for encoding the signals.
- the picture compression circuit may comprise one encoding apparatus for serially inputting a luminance signal Y and color-difference signals U, V by a block unit and for encoding the signals.
- the picture compression circuit may comprise one encoding apparatus for serially inputting color signals R, G, B by a block unit and for encoding the signals.
- the picture expansion circuit may comprise a plurality of the decoding apparatuses for inputting an encoded luminance signal Y and encoded color-difference signals U, V in parallel and for decoding the signals.
- the picture expansion circuit may comprise a plurality of the decoding apparatuses for inputting encoded color signals R, G, B in parallel and for decoding the signals.
- the picture expansion circuit may comprise one decoding apparatus for serially inputting an encoded luminance signal Y and encoded color-difference signals U, V by a block unit and for decoding the signals.
- the picture expansion circuit may comprise one decoding apparatus for serially inputting encoded color signals R, G, B by a block unit and for decoding the signals.
- Fig. 1 is a perspective view showing a configuration of a picture processing apparatus having an encoder according to the present invention.
- a picture processing apparatus having a decoder according to the present invention also has the configuration similar to that of the picture processing apparatus illustrated in Fig. 1 .
- a picture processing apparatus 60 includes a display unit 61, a keyboard 62, a mouse 63, a mouse pad 64, a system unit 65, and a compact disc drive 100.
- a picture processing apparatus having a decoder according to the present invention receives encoded picture information from the compact disc drive 100, decodes the encoded picture information, transfers the decoded picture information to the system unit 65, and displays the decoded picture information on the display unit 61.
- a picture processing apparatus according to the present invention encodes picture information displayed on the display unit 61 and supplies the encoded picture information to the compact disc drive 100.
- a picture processing apparatus according to the present invention encodes picture information and transmits the encoded picture information via a transmission line (not shown)
- the configuration of the picture processing apparatus according to the present invention does not have to be limited to the configuration of a personal computer or a workstation as shown in Fig. 1 .
- a video player for example, may be used as an input device instead of the compact disc drive 100, or picture data from a network may be input in place of the picture information.
- the input data may be either in analog format or in digital format.
- a picture processing apparatus may be provided as an independent apparatus, as shown in Fig. 1 . However, it may be housed inside the housing of a peripheral device such as a printer 66, a scanner 68, a facsimile machine 69, a display apparatus (such as the display unit 61), or a storage device (such as the compact disc drive 100), as shown in Fig. 2 . That is, a picture processing apparatus according to the present invention is defined herein to be an electronic apparatus having an encoding apparatus or a decoding apparatus which will be hereinafter described. Or, a picture processing apparatus according to the present invention is also defined to be an electronic apparatus for implementing an encoding method or a decoding method which will be hereinafter described.
- An encoding apparatus or a decoding apparatus may be provided as an independent apparatus. Alternatively, it may be incorporated into a system board of a television camera, a measuring device, or a computer, or it may be incorporated as a part of a circuit board, or it may be provided in a semiconductor chip. It may be implemented in a type of a communication system by connecting individual devices illustrated in Fig. 2 via a local area network (not shown in Fig. 2 ) to transmit encoded information therebetween. Alternatively, it may be implemented in a type of a communication system for receiving and transmitting encoded information via a wide area network such as the ISDN (Integrated Services Digital Network).
- ISDN Integrated Services Digital Network
- the encoding or the decoding method described in Fig. 67 is assumed to be applied to an encoder in an encoding apparatus or a decoder in a decoding apparatus according to the present invention.
- encoding or decoding for an extended information source of binary symbols (a binary symbol sequence) is implemented, based on the information which one of the binary symbols is MPS and based on the estimated occurrence probability of the MPS, by selecting a codeword table from a set of Huffman codeword tables prepared systematically for the extended information source of binary symbols (a binary symbol sequence) as shown in Fig. 67 .
- the code can be most suited to the state of the extended information source of binary symbols decided from the estimated occurrence probability of the MPS.
- code orders are set respectively for the thirty-two states.
- a plurality of encoders or decoders which will be hereinafter described, respectively and independently set one of the thirty-two states and one of the corresponding code orders and encode or decode binary symbols.
- a configuration and an operation of an encoding apparatus will be described explaining the switching between encoding modes and the conversion of a prediction error between the value of an encoding picture element and a prediction value therefor into a binary symbol sequence.
- Fig. 4 shows a configuration of an encoding 400 according to the first embodiment of the present invention.
- reference numeral 1 indicates a picture element memory for receiving, storing, and outputting the value of an encoding picture element.
- the picture element memory 1 also outputs the value of at least one encoded picture element which has been already stored in the picture element memory 1 and which is adjacent to the encoding picture element as a reference picture element.
- Reference numeral 2 indicates a mode determinator for deciding the mode A or the mode B, based on the value of the at least one reference picture element and for outputting a mode discrimination signal CM.
- a mode decision method for deciding the mode A or the mode B will be hereinafter described.
- Reference numeral 3 indicates a first predictor for calculating in the mode A the prediction value of the encoding picture element by referring to the value(s) of the reference picture element(s).
- Reference numeral 30 represents a first prediction error calculator for subtracting the prediction value calculated by the predictor 3 from the value of the encoding picture element and for determining the prediction error.
- Reference numeral 4 indicates a zero determinator for converting the prediction error between the value of the encoding picture element and the prediction value calculated by the predictor 3 into a binary symbol indicating whether the prediction error is 0 or not 0. Then, the zero determinator 4 outputs the converted binary symbol.
- a zero determinator is provided for determining whether a prediction error is zero or not.
- this determinator may be a determinator for determining whether or not a prediction error is a predetermined value such as, for example, one or not one, or -3 or not -3, and the like.
- Reference numeral 5 indicates a first encoder for encoding in the mode A the binary symbol which has been output from the zero determinator 4.
- Reference numeral 6 indicates a second encoder for encoding a prediction error in the mode A except when the prediction error is zero. (When the above-mentioned determinator determines whether a prediction error is one or not one, the second encoder 6 encodes a prediction error in the mode A except when the prediction error is one. Similarly, when the above-mentioned determinator determines whether a prediction error is -3 or not -3, the second encoder 6 encodes a prediction error in the mode A except when the prediction error is -3.)
- Reference numeral 7 indicates a second predictor for calculating in the mode B the prediction value of an encoding picture element by referring to the value(s) of the reference picture element(s).
- Reference numeral 31 indicates a second prediction error calculator for determining the prediction error by subtracting the prediction value calculated by the predictor 7 from the value of the encoding picture element.
- Reference numeral 8 represents a third encoder for encoding in the mode B the prediction error between the value of the encoding picture element and the prediction value calculated by the predictor 7.
- Reference numeral 9 represents a code switching device for selecting and outputting an appropriate codeword among the codewords supplied from the first encoder 5, the second encoder 6, and the third encoder 8.
- Reference numeral 10 represents a code buffer for receiving codewords selected by the code switching device 9 and for outputting a sequence of codes in the order the codewords have been supplied to the code buffer.
- Reference numeral 11 indicates an encoding controller for controlling the picture element memory 1, the first encoder 5, the second encoder 6, and the third encoder 8, the code switching device 9, and the code buffer 10, based on the mode discrimination signal CM and control signals C1 through C6.
- Reference numeral 101 indicates a first encoding section provided with the first encoder 5 which encodes in the mode A a binary symbol indicating whether or not a prediction error between the value of an encoding picture element and the estimated prediction value is zero, and the second encoder 6 which encodes a prediction error in the mode A when the prediction error is not zero.
- Reference numeral 102. indicates a second encoding section 102 provided with the third encoder 8 which encodes in the mode B a prediction error regardless of whether or not the prediction error between the value of an encoding picture element and the estimated prediction value is zero.
- Figs. 5 through 7 are block diagrams showing internal configurations of the first encoder 5, the second encoder 6, and the third encoder 8 respectively.
- the first encoder 5, the second encoder 6, and the third encoder 8 include the first probability estimator 25, the second probability estimator 26, and the third probability estimator 28 respectively for receiving a sequence of binary symbols and estimating the occurrence probability of the More Probable Symbols (MPS) for the binary symbols. Furthermore, the first encoder 5, the second encoder 6, and the third encoder 8 have a first codeword allotter 15, a second codeword allotter 16, and a third codeword allotter 18 respectively for receiving a sequence of binary symbols and the estimated occurrence probability of the MPS estimated by the probability estimators, for encoding the sequence of binary symbols, and for outputting codewords.
- MPS More Probable Symbols
- the first probability estimator 25, the second probability estimator 26, and the third probability estimator 28 determine the code orders shown in Fig. 3 , and supply the code orders to the first codeword allotter 15, the second codeword allotter 16, and the third codeword allotter 18 respectively.
- the probability estimator changes the code order to an appropriate value based on the occurrence probability of the MPS estimated from past data on binary symbol sequences. For this reason, an encoding efficiency can be enhanced.
- the first example of the two methods is the state transition method.
- the state number assigned to specify each of the thirty-two states shown in Fig. 3 is increased by one when the number of MPSs consecutively occurred has become equal to the code order.
- the state number is decreased by one when an LPS has occurred before the number of MPSs consecutively occurred becomes equal to the code order.
- the second example of the two methods is to count the number of binary symbols "0" and binary symbols "1" which have occurred in a binary symbol sequence and which are respectively indicated by N (0) and N (1), and determine the code order based on the result of the count.
- the probability estimator may estimate the occurrence probability of the MPS by using a method other than the abovementioned first code order determination method or the second code order determination method.
- the second encoder 6 and the third encoder 8 have the first error to symbol converter 36 and the second error-to-symbol converter 38 respectively for converting a prediction error to binary symbol(s).
- the error-to-symbol converter 36 performs conversion as shown in Fig. 8 .
- the error-to-symbol converter 38 performs conversion as shown in Fig. 9 .
- the conversion performed by the error-to-symbol converter 36 or the error-to-symbol converter 38 can be effected by prestoring the tables as shown in Fig. 8 and Fig. 9 and then converting a prediction error into binary symbol(s), by retrieving the tables.
- a correspondence between the prediction error and binary symbol(s) shown in Fig. 8 and Fig. 9 may be achieved by using an algorithm which will be described hereinafter.
- the algorithm involves the step of generating comparison values successively, starting with the value which would most likely occur as a prediction error to be supplied to the error to symbol converter 36 or the error-to-symbol converter 38, that is, in order from the lowest absolute value of the prediction error among those shown in Fig. 8 and Fig. 9 , and the step of comparing one by one the generated comparison values with the prediction error supplied to the error-to-symbol converter 36 or the error-to-symbol converter 38. The comparison continues until one of the generated comparison values coincides with the prediction error.
- the error-to-symbol converter 36 generates "-1" as the first comparison value. Since the first comparison value and the prediction error supplied to the error-to-symbol converter 36 do not coincide with each other, a binary symbol "0" is output. Then, the error-to-symbol converter 36 generates "+1" as the second comparison value. Since the prediction error "-2" and the comparison value "+1” do not coincide with each other, a binary symbol "0” is output again. Then, the error-to-symbol converter 36 generates the comparison value "-2". Since the prediction value and the comparison value coincide with each other this time, a binary symbol "1” is generated, and the conversion is terminated. Consequently, when the prediction error "-2" is supplied to the error-to-symbol converter 36, a binary symbol sequence "001" is output.
- Fig. 10 is a flow chart showing a mode deciding operation effected by the mode determinator 2 and a flow of an encoding operation based on the decided mode.
- the mode B is set, and the encoding mode is switched into the mode B (at S17).
- the mode A is set (at S98).
- the zero determinator 4 outputs a binary symbol "0" if a prediction error obtained by subtracting the output of the predictor 3 or the prediction value (for example, the value of the picture element which has been output immediately before the reference picture element) from the value of an encoding picture element is zero (the prediction has proved to be correct). On the other hand, if a prediction error is not zero (the prediction has proved to be incorrect), the zero determinator 4 outputs a binary symbol "1".
- the first encoder 5 implements encoding, regarding the output value "0" of the zero determinator 4 as the MPS and the output value "1" as the LPS.
- the encoding process in case that the output value of the zero determinator 4 is "0" (a prediction error is zero), and in case that the output value of the zero determinator 4 is "1" (a prediction error is not zero) proceeds as follows:
- the first encoder 5 has an MPS counter (not shown) inside the probability estimator 25.
- the MPS counter counts the number of binary symbols "0" indicating that a prediction error is zero, that is, the number of MPSs consecutively occurred.
- a one-bit codeword "0" is determined only when the number of MPSs consecutively occurred which are supplied to the first encoder 5 (the count of the MPS counter) has become equal to the code order. The codeword is not determined until the number of MPSs consecutively occurred becomes equal to the code order.
- the first encoder 5 encodes a binary symbol "1" indicating that a prediction error is not zero, that is, an LPS together with the number of MPSs which have occurred before the LPS and which have not been allotted a codeword yet.
- the codeword length is determined according to the number of MPSs which have occurred before the LPS (the count of the MPS counter).
- the code order is 2 n a codeword whose codeword length is (n+1) bits is determined.
- the second encoder 6 converts a prediction error obtained by subtracting the prediction value output of the predictor 3 from the value of an encoding picture element (-255 ⁇ - 1, 1 ⁇ 255, exclusive of zero; in the case of one picture element being represented by eight bits) into a binary symbol sequence "0...01" shown in Fig. 8 so as to implement encoding.
- the converted binary symbol sequence is divided into at least one message based on the value of the code order determined by the probability estimator 26. Then, the codeword corresponding to the at least one message is determined.
- the feature of this embodiment is that, when a prediction error is encoded by the second encoder 6, the same encoding system as that has been applied to the first encoder 5 as shown in Fig. 67 is also applied.
- a main feature of this embodiment is that, for encoding a binary symbol sequence, the same encoding method shown in Fig. 67 is applied to both the first encoder 5 and the second encoder 6.
- the predictor 7 calculates a prediction value from the value of at least one reference picture element.
- the calculation method may be implemented in accordance with a predetermined prediction function or by referring to the reference table.
- the third encoder 8 converts a prediction error (-255 ⁇ +255, inclusive of zero in the case of one picture element being represented by eight bits) obtained by subtracting the calculated prediction value from the value of an encoding picture element into a binary symbol sequence shown in Fig. 9 . Then, the third encoder 8 encodes the binary symbol sequence in the same manner as that with the second encoder 6. In other words, a codeword corresponding to the binary symbol sequence is generated, based on the encoding method illustrated in Fig. 67 . Since all of binary symbol sequences shown in Fig. 9 end with an LPS, the third encoder 8 can determine codewords corresponding to all the binary symbol sequences illustrated in Fig. 9 .
- the probability estimator 26 and the probability estimator 28 receive the binary symbol(s), change, and determine the code order based on the received binary symbol(s) and supplies the determined code order to the codeword allotter 16 and the codeword allotter 18, respectively.
- the aforementioned two code order determination methods can be applied in this case.
- Encoding effected by the codeword allotter 16 and the codeword allotter 18 are the same as encoding effected by the codeword allotter 15. It means that encoding is accomplished by using the encoding method shown in Fig. 67 .
- the encoding controller 11 directs the picture element memory 1 to store the values of encoding picture elements and output the values of the encoding picture elements and the value of at least one reference picture element adjacent to one of the encoding picture elements.
- the encoding controller operates the first encoder 5, the second encoder 6, and the third encoder 8 appropriately via the mode discrimination signal CM supplied from the mode determinator 2. During that operation process, the first encoder 5, the second encoder 6, and the third encoder 8 notify the encoding controller 11 of the state in which codewords are arranged to be output.
- the encoding controller 11 determines among the first encoder 5, the second encoder 6, and the third encoder 8 the encoder for outputting codewords and uses the code switching device 9 and the code buffer 10 to supply a code composed of a sequence of codewords in the order that the codewords have been determined.
- the code buffer 10 constitutes a code from codewords.
- Fig. 11 is a diagram showing encoding picture elements, the condition of reference picture elements, the output values of the zero determinator, and prediction errors.
- the mode A is set as the initial encoding mode for the apparatus at S11. Then, at S12, the picture element X 1 is entered. At S14, the picture element X 1 is determined to be encoded in the mode A, and at S15, the output value of the zero determinator for the picture element X 0 is checked. As shown in Fig. 11 , the output value of the zero determinator for the picture element X 0 is zero. Thus, the operation proceeds to S18. At S18, the picture element X 0 is encoded in the mode A by the first encoder 5.
- the picture element X 1 is entered. Through S14, S15, and S18, the picture element X 1 is also encoded in the mode A.
- the picture element X 2 is entered. Through S14, S15, and S18, the picture element X 2 is also encoded in the mode A.
- the picture element X 3 is entered. Then, at S14, the picture element X 3 is determined to be encoded in the mode A, and the operation proceeds to S15. At S15, since the output value of the zero determinator is 1, the operation proceeds to S17. At S17, the picture element X 3 is encoded in the mode A. Since the output value of the zero determinator is one, encoding in the mode A in this case is effected by the first encoder 5 and the second encoder 6. Since an LPS has occurred, the codeword supplied from the first encoder 5 for the picture elements X 0 , X 1 , and X 2 is determined. In Fig.
- the codeword supplied from the first encoder 5 for the picture elements X 0 , X 1 , and X 2 is completed in a position P1.
- the second encoder 6 when encoding in the mode A is determined by the occurrence of an LPS encoding by the second encoder 6 is effected for the picture element for which the output value of the zero determinator is "1".
- the prediction error of the picture element X 3 is "+1".
- the error-to-symbol converter 36 in the second encoder 6 outputs binary "01" symbols according to the table shown in Fig. 8 .
- the second encoder 6 encodes the binary symbols according to the encoding method shown in Fig. 67 .
- the mode B is set.
- the picture element X 4 is entered.
- the picture element X 4 is determined to be encoded in the mode B.
- the picture element X 4 is encoded in the mode B. Since the prediction error of the picture element X 4 is "+1", binary symbols "001" are output as shown in Fig. 9 . All the binary symbol sequences shown in Fig. 9 also end with an LPS.
- codewords supplied from the third encoder 8 are determined at the end of the encoding operation for individual prediction errors.
- the codeword for the picture element X 4 is determined in a position P3.
- the picture element X 5 is entered at S12.
- the picture element X 6 is entered at S12.
- the picture element X 6 is determined to be encoded in the mode A.
- the output value of the zero determinator for the picture element X 6 is determined to be "1". Accordingly, the picture element X 6 is encoded in the mode A at S17. Since an LPS has occurred in the case of the picture element X 6 , the codeword supplied from the first encoder 5 for the picture element X 5 is determined in a position P4, as shown in Fig. 12 . Since the first encoder 5 determines the codeword for the picture element X 5 due to an LPS occurred, the second encoder 6 encodes the prediction error for the picture element X 6 . As shown in Fig.
- the prediction error for the picture element X 6 is "-2". Accordingly, as shown in Fig. 8 , the binary "001" symbols for the prediction error of "-2" are to be encoded. Since an LPS has occurred in a position P5, the codeword supplied from the second encoder 6 for the picture element X 6 is determined in the position P5.
- all the first encoder 5, the second encoder 6, and the third encoder 8 implement encoding with the code order four.
- the first encoder 5, the second encoder 6, and the third encoder 8 respectively determine their own code orders and operates independently.
- a picture element is entered, and the code for the entered picture element is output.
- the output code is temporarily stored in a memory (not shown) in the code buffer 10. Then, the output code is transmitted from the code buffer 10 directly or indirectly via radio or a communications line in analog or digital format.
- the output code is stored in a storing medium (such as a magnetic or optical recording card, a tape, a disc, a RAM, or a ROM).
- an LPS when the encoding mode is switched from the mode A to the mode B, an LPS is always output. That is, in this first embodiment, when the encoding mode is switched from the mode A to the mode B, a codeword is always determined by the first encoder 5 or the second encoder 6. When the encoding mode is switched from the mode B to the mode A, an LPS is also always output. For this reason, in this first embodiment, when the encoding mode is switched from the mode B to the mode A, a codeword is always determined by the third encoder 8.
- this encoding apparatus provides a considerably enhanced encoding efficiency.
- Switching of encoding is effected by the encoding controller 11 which controls respective devices within the encoding apparatus, using the mode discrimination signal CM and the control signals C1 through C6.
- Fig. 13 is a block diagram showing a configuration of a decoding apparatus 500 according to the first embodiment of the present invention.
- Reference numeral 41 indicates a picture element memory for outputting at least one stored and decoded picture element adjacent to a decoding picture element as reference picture element before decoding the decoding picture element.
- the picture element memory 41 also stores decoded picture elements.
- Reference numeral 42 indicates a mode determinator, like the mode determinator 2 in the encoding apparatus described hereinbefore, for determining the decoding mode to be either the mode A or the mode B, based on the value of the at least one reference picture element. Then, the mode determinator 42 outputs the mode discrimination signal CM.
- Reference numeral 45 indicates a first decoder for decoding a codeword in the mode A into a binary symbol indicating whether or not a prediction error is zero. (When the zero determinator determines whether a prediction error is one or not one, for example, the first decoder decodes a codeword in the mode A into a binary symbol indicating whether or not a prediction error is one. Similarly, when the determinator determines whether a prediction error is -3 or not -3, the first decoder decodes a codeword in the mode A into a binary symbol indicating whether or not a prediction error is -3.)
- Reference numeral 46 indicates a second decoder for decoding a codeword in the mode A into a prediction error when the prediction error is not zero.
- Reference numeral 48 indicates a third decoder for decoding in the mode B a codeword into a prediction error between the value of a decoding picture element and the prediction value calculated by the predictor 7 for the decoding picture element.
- Reference numeral 40 indicates a code buffer for dividing an input code into codewords and outputting the codewords.
- Reference numeral 43 indicates a decoding controller for controlling the picture element memory 41, the first decoder 45, the second decoder 46, and the third decoder 48, a picture element switching device 12 (which will be described hereinafter), and the code buffer 40 based on the mode discrimination signal CM and the control signals C11 through C16.
- Reference numeral 12 indicates the picture element switching device for selecting an appropriate decoded picture element among decoded picture elements output from decoding picture element calculators 32 and 33, and the predictor 3, and outputting the value of the appropriate decoded picture element.
- Reference numerals 32 and 33 indicate the first decoding picture element calculator and the second decoding picture element calculator, respectively for calculating the value of a decoding picture element from the prediction value for the decoding picture element and the decoded prediction error.
- the predictor 3 and the predictor 7 are the same as those in the picture encoding apparatus described hereinbefore.
- Reference numeral 201 indicates a first decoding section provided with the first decoder 45 for decoding a codeword into a binary symbol indicating whether or not a prediction error between the value of a decoding picture element and the estimated prediction value is zero, and the second decoder 46 for decoding a codeword into a prediction error when the prediction error is not zero.
- Reference numeral 202 indicates a second decoding section provided with the third decoder 48 for decoding a codeword into a prediction error regardless of whether the prediction error between the value of a decoding picture element and the estimated prediction value is zero or not.
- Figs. 14, 15, and 16 are block diagrams respectively showing an internal configuration of the first decoder 45, the second decoder 46, and the third decoder 48.
- the first decoder 45, the second decoder 46, and the third decoder 48 have a first symbol restoring device 55, a second symbol restoring device 56, and a third symbol restoring device 58, respectively. Furthermore, the first.decoder 45, the second decoder 46, and the third decoder 48 have a first probability estimator 75, a second probability estimator 76, and a third probability estimator 78, respectively. The second decoder 46 and the third decoder 48 have a first symbol-to-error converter 86 and a second symbol-to-error converter 88 respectively for converting a binary symbol sequence to a prediction error.
- the first decoder 45, the second decoder 46, and the third decoder 48 receive codewords into which the code buffer 40 has divided a code and which have been supplied from the code buffer 40.
- the first decoder 45, the second decoder 46, or the third decoder 48 implements decoding with the code order 2 n
- the first decoder 45, the second decoder 46, or the third decoder 48 determine the codeword length based on the value of the starting bit of a codeword in the following way:
- the codeword length is determined to be 1. Consequently, as seen from Fig. 67 , a message comprised solely of binary symbols "0" having the codeword length 2" is restored.
- the codeword length is determined to be (n+1) bits. Consequently, the first decoder 45, the second decoder 46, or the third decoder 48 regards the value of the binary symbols for the remaining n-bit codeword except the starting bit 1 of the codeword as k or the number of binary symbols "0" consecutively occurred. Then, a message "0...01" having the codeword length of (k+1) bits is restored.
- the first decoder 45 When decoding in the mode A, the first decoder 45 receives a codeword, converts the codeword into a binary symbol sequence, and outputs the binary symbol sequence (which is equivalent to a message). Each output of the first decoder 45 is for a picture element which is to be decoded in the mode A. when the output value of the first decoder 45 is "0", it indicates that a prediction error for a decoding picture element is zero. On the other hand, when the output value of the first decoder 45 is "1", it indicates that a prediction error is not zero. When a prediction error is zero, the prediction value will become the value of a decoding picture element. When a prediction error is not zero, the prediction error is subsequently decoded by the second decoder 46.
- the second decoder 46 and the third decoder 48 respectively receive codewords and convert the codewords into at least one message. Then, the second decoder 46 or the third decoder 48 combine the at least one message into binary symbol sequences.
- the second decoder 46 implements decoding in the mode A for a picture element for which the prediction error is not zero.
- the second decoder 46 inversely converts a binary symbol sequence, which is as shown in Fig. 8 (which indicates the case that a prediction error of zero does not exist), into a prediction error and outputs the prediction error.
- the third decoder 48 implements decoding for a picture element which has been encoded in the mode B.
- the third decoder 48 inversely converts a binary symbol sequence, which is as shown in Fig. 9 (which indicates the case that a prediction error of zero exists), into a prediction error and outputs the prediction error.
- the value of a decoding picture element which is to be decoded in the mode A is obtained by adding the output of the predictor, 3 (the prediction value) to the output of the second decoder 46 (the prediction error).
- the value of a decoding picture element which is to be decoded in the mode B is obtained by adding the output of the predictor 7 (the prediction value) to the output of the third decoder 48 (the prediction error).
- Mode switching in a decoding process can be implemented by the mode determinator 42, by using the same determination method as with the mode determinator 2 shown in Fig. 4 , based on.the condition of reference picture element(s) output from the picture element memory 41 and the output value of the first decoder 45.
- the decoding controller 43 directs the picture element memory 41 to output at least one reference picture element, and also operates the first decoder 45, the second decoder 46, and the third decoder 48 selectively according to the mode discrimination signal CM output from the mode determinator 42.
- the first decoder 45, the second decoder 46, and the third decoder 48 notify the decoding controller that a codeword has been input.
- the code buffer 40 divides a code into codewords.
- the decoding controller 43 determines the appropriate decoder for implementing decoding among the first decoder 45, the second decoder 46, and the third decoder 48 and directs the picture element switching device 12 to supply decoded picture elements to the picture element memory 41 for storing them.
- Fig. 17 is a block diagram showing a configuration of an encoding apparatus 400 according to the second example.
- Fig. 17 is different from Fig. 4 described in the first embodiment, in that the output of the zero determinator 4 is'riot supplied to the mode determinator 2.
- the picture element X is encoded in the mode A (at S4 through S6).
- the picture element X is encoded in the mode B (at S7).
- the output value of the zero determinator 4 is referred to as in the first embodiment, Since encoding operations in the mode A and the mode B are same as those in the first embodiment, the description about the encoding operations will be omitted.
- a codeword is always determined by the first encoder 5.
- a codeword is not always determined by the first encoder 5.
- Fig. 19 shows an example of binary symbols into which picture elements X 1 through X 6 are encoded according to the second example of the present invention.
- the example shown in Fig. 19 shows the encoding state where the encoding picture elements shown in Fig. 11 are input into the encoding apparatus according to this example.
- a codeword is supplied from the first encoder 5, the second encoder 6, and the third encoder 8 for implementing encoding with the code order four..
- a codeword is supplied from the first encoder 5, the second encoder 6, and the third encoder 8 for implementing encoding with the code order four.
- the example shown in Fig. 19 at the time of termination of the encoding for the picture element X 2 in the mode A, binary symbols output consecutively have not ended with 1.
- the count of the MPS counter has not reached the code order four, either.
- a codeword for the picture element X 2 is not determined by the first encoder 5.
- codewords for the picture elements X 3 and X 4 are determined by the third encoder 8.
- the decoding apparatus 500 cannot implement accurate decoding unless a codeword generated by the first encoder 5 is input into the decoding apparatus before a codeword generated by the third encoder 8 is input into the decoding apparatus.
- an operation for changing the transmission order of codewords which will be described hereinafter, that is, control over the transmission order of codewords is required in the encoding apparatus 400.
- a picture-element preread operation in the mode A which will be described hereinafter, is required in the encoding apparatus 400.
- Fig. 20 is a flow chart showing a flow of the control process implemented by the encoding controller 11 after a codeword has been determined by the third encoder 8 (at S20).
- Fig. 21 is a flow chart showing a flow of the control process implemented by the encoding controller 11 after a codeword has been determined by the first encoder 5 (at S30).
- a codeword is determined by the third encoder 8.
- the codeword generated by the third encoder 8 is temporarily stored in the code buffer 10 (at S24 in Fig. 20 ).
- a codeword is determined by the first encoder 5:
- codewords are determined by the third encoder 8 in positions P1 and P2. While on the other hand, a codeword which is to be output from the first encoder 5 is determined in a position P3. Consequently, the codewords determined by the third encoder 8 in the positions P1 and P2 are temporarily stored in the code buffer 10. When a codeword to be output from the first encoder 5 has been determined in the position P3, the codeword determined in the position P3 is output first. Then, the codewords temporarily stored in the code buffer 10 and determined in the positions P1 and P2 are output subsequently. Finally, a codeword determined by the second encoder 6 in a position P4 is output.
- a codeword determined by the first encoder 5, a codeword determined by the second encoder 6, and a codeword determined by the third encoder 8 and temporarily stored in the code buffer 10 can be transmitted in stated order.
- the output order of codewords shown in Fig. 19 is such that a codeword determined by the first encoder 5 in the position P3 is output first. Then, a codeword determined by the second encoder 6 in the position P4 is output. Then, a codeword determined by the third encoder 8 in the position P1 is output. Finally, a codeword determined by the third encoder 8 in the position P2 is output.
- a maximum capacity MAX of the code buffer 10 needs to be determined both by the transmitting side and the receiving side.
- Fig. 23 shows an operation for defining the maximum capacity MAX of the code buffer 10.
- dummy binary symbols "0" whose number is equal to the number required for determining a codeword in the mode A should be added to binary symbols for which a codeword has not been determined. Since Fig. 23 shows a case where the code order is four, two dummy binary symbols "0" are added. Then, the codeword to be output from the first encoder 5 is determined. Dummy binary symbols "0" are added to binary symbols for which a codeword has not been determined by the required number until the codeword is determined by the first encoder 5.
- an accurate decoding operation can be effected by setting the maximum capacity MAX of the code buffer in the decoding apparatus 500 to be identical to that in the encoding apparatus 400 and by implementing a process of adding dummy binary symbols "0".
- Fig. 23 shows a case where dummy binary symbols "0" are added to binary symbols for which a codeword has not been determined so as to determine a codeword to be output from the first encoder 5.
- a codeword to be output from the first encoder 5 may be determined by adding only a dummy binary symbol "1" to the binary symbols. In this case, however, a codeword for a prediction error should not be generated from the second encoder 6 when a dummy binary symbol "1" has been added.
- picture element preread operation in the encoding mode A is employed to mean an operation for reading the values of picture elements to be encoded subsequently one by one beforehand until a codeword to be output from the first encoder 5 is determined.
- Fig. 24 is a flow chart showing a flow of the control process initiated by the encoding controller 11 after the mode B has been set (at S50).
- the picture element preread operation in the encoding mode A after the mode B has been set at S50 as shown in Fig. 24 , whether or not the count of the MPS counter inside the first encoder 5 is zero is checked at S52.
- the count of the MPS counter is zero, it means that a codeword to be output from the first encoder 5 has already been determined.
- the picture element preread operation is not required.
- the count of the MPS counter inside the first encoder 5 is not zero, it means that a codeword to be output from the first encoder 5 has not been determined. For this reason, the picture element preread operation shown at S53 through S56 will be implemented.
- the values of the picture elements X 5 and X 6 are preread so as to determine codewords to be output from the fist encoder 5 (at S55). Thereafter, the picture elements X 3 and X 4 which have not been encoded yet are encoded in the mode B (at S57).
- the values of the picture elements to be encoded subsequently are to be preread. Thereafter, encoding should be implemented for the picture elements which have not been encoded yet due to the picture element preread operation. For this reason, a line memory for storing picture elements is needed, and both the transmitting side and the receiving side need to predefine the maximum capacity of the line memory.
- the reason for predefining the maximum capacity of the line memory is the same as that given from Fig. 23 . That is, because, even if picture elements to be encoded subsequently are preread for a codeword: undetermined by the first encoder 5, it occurs that a codeword may still be undetermined by the first encoder 5. Thus, the maximum capacity of the line memory for storing picture elements should be predefined. If the number of picture elements is to exceed the maximum capacity of the line memory, dummy binary symbols "0" are added to binary symbols for which a codeword has not been determined yet so as to determine the codeword to be output from the first encoder 5. As described hereinbefore, a dummy binary symbol "1" may be added to determine the codeword to be output from the first encoder 5.
- Fig. 25 is different from Fig. 13 used in the first embodiment in that the output of the first decoder 45 is not supplied to the mode determinator 42.
- the first decoder 45 shown in Fig. 25 implements the same decoding operation as that of the first decoder 45 shown in Fig. 13 according to the first embodiment. That is, the first decoder 45 decodes a codeword in the mode A into a binary symbol indicating whether or not a prediction error is zero. Then, the result is transmitted to the decoding controller 43 through the control signal C13.
- the decoding controller 43 operates the second decoder 46 based on the information from the first decoder 45 as to whether or not the prediction error is zero.
- a description will be especially directed to an operation different from that according to the first embodiment of the present invention.
- the decoding apparatus 500 converts an input codeword into a binary symbol sequence (or a message). Using such messages, the values of picture elements are reproduced.
- the decoding mode is switched from the mode A into the mode B before all of the binary symbols into which the first decoder 45 decodes one codeword in the mode A are used up. Referring again to Fig. 19 , when decoding X 1 binary symbols " 0001" associated with decoding of X 2 , X 5 , and X 6 can be obtained. However, after X 2 is decoded, the decoding mode is switched into the mode B. Thus, binary symbols "01" remain unused.
- the decoding controller 43 directs the first decoder 45 to output the unused binary symbols.
- the decoding controller 43 directs the first decoder 45 to receive a codeword, converts it into a sequence of binary symbols, and decodes the binary symbols.
- the encoding mode for a picture element can be determined solely by the state to which reference picture elements belong. Consequently, a codeword may not be determined when the encoder is switched from the first encoder 5 to the third encoder 8.
- the decoding apparatus 500 can decode codewords into picture elements consistently.
- a codeword is always determined as long as binary symbols shown in Fig. 9 are encoded according to the encoding method shown in Fig. 67 .
- control over the transmission order of codewords and control over the picture element preread operation are not needed.
- Fig. 26 shows a configuration of an encoding apparatus 400 according to the third example.
- Fig. 26 is different from Fig. 17 described in the second example in that the encoders are configured in a different way.
- Figs. 27, 28, and 29 are block diagrams showing a configuration of the first encoder 5a, the second encoder 6a, and the third encoder 8a respectively according to this example.
- the first encoder 5a, the second encoder 6a, and the third encoder 8a have exclusive OR circuits 95, 96, and 98 respectively.
- Each of the exclusive OR circuits 95, 96, and 98 receives a binary symbol X and an MPS (Y) supplied from probability estimators 25a, 26a, and 28a respectively, performs an operation as shown in a truth table in Fig. 30 , and outputs an exclusive OR signal Z to the probability estimator.
- the MPS (Y) used herein is defined as the value of a binary symbol which the probability estimators interpret as the MPS at the time of an encoding operation, that is, "0" or "1".
- the third example is different from the second example in that, when encoding is implemented according to the encoding method shown in Fig. 67 , interpretation of the MPS and the LPS is changed during transition of states.
- the first encoder 5 interprets the MPS as "0" (a binary symbol indicating that the prediction has proved to be correct).
- the second encoder 6 and the third encoder 8 interpret the MPS as "0" defined in Fig. 8 and Fig. 9 .
- the third example as shown in a flow chart illustrated in Fig.
- the exclusive OR signal Z which is the output of the exclusive OR circuit is regarded to represent a binary symbol for determining the transition of states described in the first embodiment and second example.
- the transition of states can be implemented based on the exclusive OR signal Z.
- Described above is the case where the interpretation of the MPS and the LPS is changed by using the state transition method which has been described as the first example of the methods of determining the code order.
- the change in the interpretation of the MPS and the LPS can also be implemented easily by comparing N(0).and N(1) in the 0/1 counting method which has been described hereinbefore as the second example of the methods of determining the code order. That is, if N(0) > N(1), the MPS may be interpreted as "0", and the LPS may be interpreted as "1". On the other hand, if N(0) ⁇ N(1), the MPS may be interpreted as "1", and the LPS may be interpreted as "0".
- the method of controlling the transmission order of codewords will be different from that described in the second example.
- the third encoder 8 which has been switched from the first encoder 5
- control over the transmission order of codewords becomes possible only if it is found whether or not a codeword has been determined.
- a codeword is not always completed in the codeword completion position shown in Fig. 19 .
- Fig. 31 shows how encoding is implemented when the interpretation of the MPS and the LPS has been changed.
- the binary symbols X is converted to the exclusive OR signal Z by the exclusive OR circuit.
- the encoders 5a, 6a, and 8a should receive a plurality of exclusive OR signals Z as a sequence of binary symbols and encode them.
- the binary symbols X constitute a binary symbol sequence which always ends with an LPS.
- the exclusive OR signal Z is the inverse of the binary symbols X, the binary symbols shown in Fig. 8 and Fig. 9 are to be inverted.
- a codeword may not always be determined by each of the encoders.
- the encoding controller 11 needs to know whether or not a codeword has been determined by the encoder which has been used before switched into the other encoder.
- the encoding controller 11 examines if there remains the other encoder by which a codeword has not been determined (or the encoder with the count of the MPS counter therein being equal to or greater than one). If there remains no such an encoder, the encoding controller 11 directs the code buffer 10 to output the determined codeword. If there remains the encoder by which a codeword has not been determined, the codeword determined by the encoder 8a is temporarily stored in the code buffer 10. Then, the codeword determined by the encoder 8a is output after another codeword has been determined by the other encoder. As shown in Fig.
- a codeword to be output from the encoder 5a has not been determined yet in the position P1.
- the codeword determined in the position P1 is not output until the codeword to be output from the encoder 5a has been determined in a position P2.
- a codeword to be output from the encoder 8a cannot be determined.
- the codeword to be output from the encoder 5a is determined in the position P2, and output as described hereinbefore, before the codeword determined in the position P1 is output. Then, a codeword determined by the encoder 6a in a position P3 is temporarily stored in the code buffer 10.
- a codeword to be output from the encoder 8a is determined in the position P4. Consequently, the codeword determined in the position P3 is output after the codeword determined in a position P4 is output.
- Figs. 33, 34, and 35 are block diagrams showing an inside configuration of decoders 45a, 46a, and 48a respectively.
- the decoding apparatus 500 converts an input codeword into a sequence of binary symbols (message), and by using the binary symbols, the value of the corresponding picture element is reproduced.
- a decoder before binary symbols decoded from a codeword by a decoder are used up, there is a case that binary symbols which have been decoded by the other decoder is used.
- the decoding controller 43 directs the decoder to output the unused binary symbols.
- the decoding controller 43 directs the decoder to receive another codeword and decode it into binary symbols.
- control over the order in which decoded binary symbols are to be used is exercised differently from that in the second example. Since other operations are performed in the same way as in the second-example, the description about the operations will be omitted.
- Fig. 36 is a block diagram showing a configuration of an encoding apparatus 400 according to the fourth embodiment of the present invention.
- encoders 5a, 6a, and 8a are encoders having the same configuration as that shown in Figs. 27, 28, and 29 , respectively.
- the fourth embodiment is different from the first embodiment in that when encoding is implemented according to the encoding method shown in Fig. 67 , interpretation of the MPS and the LPS is changed during transition of states.
- the MPS is interpreted as a binary symbol "0" (which indicates that the prediction has proved to be correct).
- the MPS is interpreted as "0" defined in Fig. 8 and Fig. 9 .
- interpretation of the MPS and the LPS is changed when an LPS has occurred in the state S 0 at the time of the transition of states.
- the rule for the transition of states is applied in the same way as in the first embodiment and the second and third examples. As described hereinbefore, when implementing transition of states by the 0/1 counting method, the change in the interpretation of the MPS and the LPS can be implemented easily by comparing N(0) with N(1).
- the method of controlling the transmission order of codewords will be the same as that in the third example.
- control over the transmission order of codewords becomes possible when it is found whether or not a codeword has been determined only when the encoder is switched from the first encoder 5 to the third encoder 8.
- codewords are not always completed in the codeword completion positions P1 through P4 shown in Fig. 19 . For this reason, whenever one encoder has been switched into the other encoder, the encoding controller should know whether or not a codeword has been determined by the encoder which has been used before switched into the other encoder.
- decoders 45a, 46a, and 48a have the same configuration as that of the decoders shown in Figs. 33, 34, and 35 respectively.
- the decoding apparatus 500 converts an input codeword into a sequence of binary symbols (message), and by using the binary symbols, the value of a picture element is reproduced.
- the decoding controller directs the decoder to output the unused binary symbols.
- the decoding controller directs the decoder to receive another codeword and decode it into binary symbols.
- control over the order in which decoded binary symbols are to be used is implemented in a different manner from that in the third example. Since other operations are performed in the same way as in the third example, the description about the operations will be omitted.
- each encoder has an MPS counter, a state-number memory for storing a state number, and an MPS memory for storing the value of MPSs inside.
- MPS counters there may be a plurality of MPS counters, state-number memories, and MPS memories inside the encoder.
- MPS counters C1, C2, C3..., state-number memories R1, R2, R3..., and MPS memories Q1, Q2, Q3... are provided corresponding to the states to which reference picture elements belong.
- the decoder can be used in the aforementioned embodiments and examples.
- Fig. 39 is a block diagram showing another configuration of an encoding apparatus 400 according to the this example.
- the predictor 3 and the predictor 7 described hereinbefore calculates the prediction value of an encoding picture element by using the same prediction method
- the predictor 3 and the predictor 7, the first prediction error calculator 30 and the second prediction error calculator 31 need not be provided separately.
- the output of the prediction error calculator 30 including the output of the predictor 3 may be supplied to the third encoder 8.
- Fig. 40 is a block diagram showing a configuration of a decoding apparatus 500 for the encoding apparatus 400 shown in Fig. 39 .
- Fig. 40 as in Fig. 39 , only one predictor and one decoding picture element calculator are provided in the decoding apparatus 500.
- Fig. 41 is a block diagram showing another configuration of the encoding apparatus 400 according to this embodiment.
- Fig. 42 is a block diagram showing an internal configuration of an encoder 5b shown in Fig. 41 .
- the feature of the encoding apparatus 400 shown in Fig. 41 is that the aforementioned first encoder 5 and the second encoder 6 are combined into the encoder 5b.
- the configuration of the encoder 5b is as shown in Fig. 42 .
- the encoder 5b has a switch 85 which connects the zero determinator 4 to the probability estimator 25 based on the control signal C2.
- the switch 85 also connects the error-to-symbol converter 36 and the probability estimator 25 based on the control signal C3.
- the first encoder 5 and the second encoder 6 receive different inputs, but use the same encoding method for encoding binary symbols. For this reason, by using the encoder 5b as shown in Fig. 42 , the configuration of an encoding apparatus can be simplified.
- Fig. 43 is a block diagram showing a decoding apparatus 500 for the encoding apparatus 400 shown in Fig. 41 .
- Fig. 44 is a block diagram showing an internal configuration of a decoder 45b used in the decoding apparatus 500 shown in Fig. 43 .
- the first decoder 45 and the second decoder 46 described in the aforementioned embodiments and examples are combined into the decoder 45b shown in Fig. 44 .
- Fig. 45 is a block diagram showing another configuration of an encoding apparatus 400 according to this example.
- Fig. 46 is a block diagram showing a configuration of an encoder 6b in the encoding apparatus 400 shown in Fig. 45 .
- Fig. 47 is a block diagram showing a decoding apparatus 500 for the encoding apparatus 400 shown in Fig. 45 .
- Fig. 48 is a block diagram showing a configuration of a decoder 46b in the decoding apparatus 500 shown in Fig. 47 .
- the feature of the encoding apparatus 400 shown in Fig. 45 is that the second encoder 6 and the third encoder 8 are combined into the encoder 6b.
- the only difference between the second encoder 6 and the third encoder 8 described in the aforementioned embodiments and examples is that the correspondence between the prediction errors and the binary symbols shown in Figs. 8 and 9 becomes different. Except for this, the encoder 6b implements the same operations.
- the second encoder 6 and the third encoder 8 can be combined into the encoder 6b.
- the second decoder 46 and the third decoder 48 shown in Fig. 13 or Fig. 25 can be combined into the decoder 46b.
- Fig. 49 is a block diagram showing another configuration of the encoding apparatus 400 according to this example.
- Fig. 50 is a block diagram showing a encoder 5c into which the first encoder 5, the second encoder 6, and the third encoder 8 described in the aforementioned embodiments and examples are combined.
- the switch 85 switches between the output of the zero determinator, the output of the error-to-symbol converter 36, and the output of the error-to-symbol converter 38.
- the encoder 5c can have the functions of the first encoder 5, the second encoder 6, and the third encoder 8 described hereinbefore.
- the first decoder 45, the second decoder 46, and the third decoder 48 can be combined into one decoder in the same way as with the encoder 5c.
- Fig. 51 is a block diagram showing another configuration of an encoding apparatus 400 according to this example.
- the encoding apparatus 400 shown in Fig. 51 does not have the mode determinator 2 and the second encoding section 102 provided in the encoding apparatus 400 illustrated in Fig. 4 .
- the result of a determination which is supplied from the zero determinator 4 to the mode determinator 2 is supplied to the encoding controller 11.
- the encoding controller 11 uses the control signals C2 and C3 to operate the first encoder 5 and the second encoder 6.
- the encoding controller 11 uses the control signal C5 to operate the code switching device 9 to select an appropriate codeword from the codewords supplied from the first encoder 5 and the second encoder 6 and supplies the appropriate codeword.
- Fig. 52 is a block diagram showing another configuration of a decoding apparatus 500 according to this example.
- the decoding apparatus 500 illustrated in Fig. 52 does not have the mode determinator 42 and the second decoding section 202 provided in the decoding apparatus 500 illustrated in Fig. 13 .
- a codeword is decoded into a binary symbol representing a result of determination which indicates whether or not a prediction error is zero by the first decoder 45, and the result of the determination is supplied to the decoding controller 43.
- the decoding controller 43 uses the control signal C12 to control the picture element switching device 12.
- the feature of the encoding apparatus 400 illustrated in Fig. 51 is that an encoding picture element is encoded based on the result of a determination as to whether the prediction for the encoding picture element has been correct or incorrect.
- the feature of the decoding apparatus 500 illustrated in Fig. 52 is that a codeword is decoded into a decoding picture element corresponding thereto based on the result of a determination as to whether the prediction for the value of the decoding picture element has been correct or incorrect.
- the predictor 3 and the predictor 7 calculates a prediction value of an encoding picture element from the value (s) of reference picture element (s). If necessary, a prediction value may also be calculated based on the mode discrimination signal CM, which can be supplied to the predictor 3 and the predictor 7 from the mode determinator, and the value(s) of reference picture element(s).
- reference picture elements are assumed to be three picture elements a, b, and c in the above-mentioned embodiments and examples, no less than one picture element may constitute reference picture element.
- all or a part of the functions of the aforementioned configured elements may be implemented by computer hardware.
- the functions of the configured elements may be implemented by software, or by firmware.
- the functions of the configured elements may be implemented.
- all or a part of the functions of the configured elements may be stored and implemented in a semiconductor chip.
- Fig. 53 shows a picture processing apparatus (a digital camera) as an embodiment of the picture processing apparatus according to the present invention.
- the figure shows an example of a configuration of a digital camera 600, which writes a static picture on a memory card.
- a picture pick-up device 325 picks up a picture composed of a plurality of picture elements
- a charge coupled device (CCD) 311 is used for a sensor which converts an optical picture into an electric signal, and the optical picture is formed on the CCD 311 through a lens 312.
- a diaphragm 313 regulates incident dose
- a shutter 314 adjusts an exposure time of the CCD 311
- an AD (Analog to Digital) converter 315 converts an output of the CCD 311 into a digital signal.
- a frame memory 316 being a semiconductor memory, temporarily stores one frame of digital signals converted from the CCD 311.
- a write signal processing circuit 317 selects at least two adjacent lines of data from the output signals of the CCD 311 stored in the frame memory 316.
- the write signal processing circuit 317 generates a luminance signal Y, color-difference signals U and V by an operation of signals of each picture elements and outputs them.
- a picture compression circuit 318 encodes and compresses the data of the luminance signal Y, the color-difference signals U, and V generated by the write signal processing circuit 317.
- a memory card 319 is configured by a semiconductor memory and stores the picture.
- a picture expansion circuit 320 expands the encoded picture data read from the memory card 319.
- a read signal processing circuit (or display circuit) 321 displays the extended luminance signal Y, and the extended color-difference signals U, V on a monitor 324.
- a system controller 322 controls an operation of the CCD 311.
- a trigger switch 323 starts picking up a picture.
- Fig. 54 shows a sequence of writing procedure of a static picture by the digital camera 600 shown in Fig. 53 .
- the trigger switch 323 is turned ON and a static picture writing sequence is started as will be described hereinafter.
- dark change accumulated in the CCD 311 is removed.
- a photometric operation is performed by a photometric element (not shown) to set an appropriate exposure time and an appropriate exposure diaphragm.
- the shutter 314 is opened and an exposure operation is performed to the CCD 311.
- the shutter 314 is closed and charge of the exposure signal is read from the CCD 311.
- the signal read from the CCD 311 is converted by the AD converter 315 into digital signal.
- One frame of the digital signals is temporarily stored in the frame memory 316.
- the reading operation of the exposure signal is finished.
- the signal temporarily stored in the frame memory 316 is read.
- the luminance signal Y and the color-difference signals U, V are obtained from the operation of a plurality of adjacent picture elements of the CCD 311 by the write signal processing circuit 317.
- the data of the luminance signal Y and the color-difference signals U, V are encoded, compressed and stored in the memory card 319.
- Fig. 55 shows a configuration of the picture compression circuit 318.
- Three encoding apparatuses 400 are provided to the picture compression circuit 318.
- the encoding apparatuses described in the above first through sixth embodiments and examples may be used for the encoding apparatuses 400.
- the encoding apparatus 400 shown in Fig. 4 can be applied.
- the encoding apparatuses 400 input the luminance signal Y and the color-difference signals U, V in parallel from the write signal processing circuit 317.
- Each encoding apparatus implements encoding by using the aforementioned encoding method.
- the encoded result is output to the memory card 319 and stored there.
- the luminance signal Y is one frame of luminance signals Y, having Y1, Y2, ..., Yn.
- the encoding apparatus 400 inputs one frame of the luminance signals Y and encode Y1, Y2, ..., Yn.
- the color-difference signals U and V are encoded by each encoding apparatus as well.
- Fig. 56 shows a configuration of the picture expansion circuit 320.
- Three decoding apparatuses 500 are provided to the picture expansion circuit 320.
- the decoding apparatuses described in the above first through sixth embodiments and examples may be used for the decoding apparatuses 500.
- the decoding apparatus 500 shown in Fig. 13 can be applied.
- the picture expansion circuit 320 inputs the luminance signal Y and the color-difference signals U, V in parallel from the memory card 319.
- the luminance signal Y and the color-difference signals U, V are decoded by each decoding apparatus 500, respectively, and output to the read signal processing circuit 321.
- Fig. 57 shows another example of the picture compression circuit 318.
- one encoding apparatus 400 is provided to the picture compression circuit 318.
- the encoding apparatus 400 inputs the luminance signal Y and the color-difference signals U, V serially frame by frame from the write signal processing circuit 317. That is, one frame of the luminance signals Y, having Y1, Y2, ..., Yn is input, and one frame of the color-difference signals U, having U1, U2, ..., Un is input, and then, one frame of the difference signals V, having V1, V2, ..., Vn, is input.
- one encoding apparatus 400 suffices for the above operation, consequently, a configuration of the picture compression circuit 318 is simplified.
- Fig. 58 shows another configuration of the picture expansion circuit 320.
- the picture expansion circuit 320 receives codes encoded by a frame unit from the memory card 319 serially.
- the picture expansion circuit 320 decodes the luminance signal Y and the color-difference signals U, V by a frame unit to output to the read signal professing circuit 321.
- signals are encoded or decoded by a frame unit.
- Signals can be encoded or decoded by a certain size of block other than a frame unit. Further, it is possible to encode or decode by a plurality of lines.
- Fig. 59 shows another configuration of the digital camera 600.
- the digital camera shown in Fig. 59 displays the picture on the monitor 324 through the frame memory 316.
- Fig. 60 shows yet another configuration of the digital camera 600.
- the digital camera shown in Fig. 60 does not include the memory card 319 and the picture compression circuit 318 and the picture expansion circuit 320 are located before and after the frame memory 316, respectively in the digital camera.
- Fig. 61 shows another configuration of the digital camera 600 and the computer 700.
- the picture compression circuit 318 of the digital camera 600 implements compression process and the computer 700 implements expansion process.
- the digital camera 600 only compression is implemented and the compressed data is stored in the memory card 319.
- the memory card 319 is coupled to the computer 700 off-line.
- a CPU (Central Processing Unit) 701 and a picture expansion program 703 stored in a RAM (Random Access Memory) 702 read the code stored in the memory card 319 and expands the read code. Then, required processing of the picture is performed by a picture processing program 704 and the picture can be displayed or printed out. If communication function is provided to the digital camera 600, the code stored in the memory card 319 can be transmit to a distant place through wired communication or wireless communication.
- Figs. 62 and 63 show cases where the picture compression circuit 318 inputs the color signals R, G, and B.
- Figs. 64 and 65 show cases where the picture compression circuit 318 inputs the color signals R, G1, G2 and B.
- the picture expansion circuit 320 decodes the color signal by a configuration similar to the configurations of Figs. 62 through 65 , which is not shown in the figure.
- the codes are accumulated by the memory card 319 .
- Secondary storage device such as a flexible disk, a harddisk, a flash memory can be applied instead of the memory card 319.
- the code is not stored in the storage device, but transmitted to the outside by a communication apparatus or cable and so on.
- the present invention provides an efficient encoding apparatus, an efficient decoding apparatus, an efficient encoding method, and an efficient decoding method. More specifically, according to the present invention, picture information is encoded or decoded efficiently in accordance with the encoding methods corresponding to a plurality of encoding modes.
- picture information can be appropriately encoded and correctly decoded even if a code is not determined at the time of switching between a plurality of encoding modes.
- the encoding apparatus or the decoding apparatus can be constructed in a compact size.
- a larger amount of picture information can be stored in the recording medium provided to the picture processing apparatus such as a digital camera than the conventional picture processing apparatus.
- the present invention has sufficient industrial applicability.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
- Compression Of Band Width Or Redundancy In Fax (AREA)
- Image Processing (AREA)
Claims (10)
- Kodiervorrichtung (400), umfassend:einen Bildelementspeicher (1) zum Empfangen und Speichern eines Wertes eines zu kodierenden Kodierbildelementes und zum Ausgeben von Werten von kodierten Bildelementen benachbart zu dem Kodierbildelement als Werte von Referenzbildelementen;eine Modus-Bestimmeinheit (2) zum Bestimmen einer spezifischen Kodierart oder einer von der spezifischen Kodierart abweichenden Kodierart aus einer Vielzahl von vorbestimmten Kodierarten für das Kodierbildelement;eine erste Kodiereinheit (101) zum Voraussagen des Wertes des Kodierbildelementes, Bestimmen eines Voraussageergebnisses, ob die Voraussage korrekt ist oder nicht, Kodieren des Wertes des Kodierbildelementes basierend auf dem Voraussageergebnis der Bestimmung und zum Ausgeben eines Codewortes, das das Kodierbildelement bei der festgelegten spezifischen Kodierart repräsentiert;eine zweite Kodiereinheit (102) zum Voraussagen des Wertes des Kodierbildelementes, Kodieren des Wertes des Kodierbildelementes und zum Ausgeben eines Codewortes, das das Kodierbildelement bei der von der spezifischen Kodierart abweichenden Kodierart repräsentiert;eine Kodiersteuereinheit (11) zum ausgewählten Betreiben der ersten Kodiereinheit und der zweiten Kodiereinheit basierend auf einer von der Modus-Bestimmeinheit (2) bestimmten Kodierart,wobei die erste Kodiereinheit (101) umfasst:eine erste Voraussageeinheit (3) zum Berechnen eines Voraussagewertes des Kodierbildelementes basierend auf Werten der Referenzbildelemente;einen ersten Voraussagefehler-Berechner (30) zum Berechnen einer Differenz zwischen dem Wert des Kodierbildelementes und dem von der ersten Voraussageeinheit (3) berechneten Voraussagewert als Voraussagefehler;eine Bestimmvorrichtung (4) zum Bestimmen, ob der von dem ersten Voraussagefehler-Berechner (30) berechnete Voraussagefehler angibt, dass die Voraussage korrekt ist oder nicht, und zum Ausgeben des Voraussageergebnisses der Bestimmung;einen ersten Kodierer (5) zum Empfangen und Kodieren mit einer Kodierfolge des Voraussageergebnisses der von der Bestimmvorrichtung (4) ausgegebenen Bestimmung und zum Ausgeben eines Codewortes, das die Zahl der eine korrekte Voraussage angebenden Voraussageergebnisse repräsentiert, die vor einem Voraussageergebnis aufgetreten sind, das angibt, dass die Voraussage nicht korrekt ist, oder zum Ausgeben eines Codewortes, wenn die Zahl gleich der Codefolge wird; undeinen zweiten Kodierer (6) zum Kodieren des Voraussagefehlers, der von dem ersten Voraussagefehler-Berechner (30) berechnet wurde, in das Codewort, wenn die Voraussage nicht korrekt ist, und zum Ausgeben des das Kodierbildelement repräsentierenden Codewortes; undwobei die zweite Kodiereinheit (102) umfasst:eine zweite Voraussageeinheit (7) zum Berechnen eines Voraussagewertes des Kodierbildelementes basierend auf Werten der Referenzbildelemente;einen zweiten Voraussagefehler-Berechner (31) zum Berechnen einer Differenz zwischen dem Wert des Kodierbildelementes und dem von der zweiten Voraussageeinheit (7) berechneten Voraussagewert als einen Voraussagefehler;einen dritten Kodierer (8) zum Kodieren des von dem zweiten Voraussagefehler-Berechner (31) berechneten Voraussagefehlers in das Codewort und zum Ausgeben des das Kodierbildelement repräsentierenden Codewortes; undwobei die Modus-Bestimmeinheit (2) die Werte der Referenzbildelemente und das von der Bestimmvorrichtung (4) ausgegebene Voraussageergebnis der Bestimmung empfängt und eine Kodierart zwischen der spezifischen Kodierart (Mode A), die von der ersten Kodiereinheit (101) verwendet wird, und der von der spezifischen Kodierart abweichenden Kodierart (Mode B), die von der zweiten Kodiereinheit (102) verwendet wird, auswählt, derart, dass die spezifische Kodierart (Mode A) fortgesetzt wird, wenn die Bestimmvorrichtung (4) bestimmt, dass die Voraussage sich als korrekt erwiesen hat, wenn die Bestimmvorrichtung (4) bestimmt, dass die Voraussage sich als nicht korrekt erwiesen hat, das Kodierbildelement in der spezifischen Kodierart (Mode A) kodiert wird und dann die Kodierart (Mode B) abweichend von der spezifischen Kodierart festgelegt wird, wobei die spezifische Kodierart (Mode A) gewählt wird, wenn die Werte der Referenzbildelemente gleich sind, und die von der spezifischen Kodierart abweichende Kodierart (Mode B) weitergeführt wird, wenn die Werte der Referenzbildelemente nicht gleich sind.
- Kodiervorrichtung (400) nach Anspruch 1, bei der der erste Kodierer (5) eine erste Wahrscheinlichkeitsschätzeinheit (25) zum Empfangen des Voraussageergebnisses der Bestimmung, das von der Bestimmvorrichtung (4) als eine Sequenz von Binärsymbolen ausgegeben wird, und zum Schätzen einer Wahrscheinlichkeit des Auftretens eines der Binärsymbole und einen ersten Codewort-Zuordner (15) zum Kodieren der Sequenz von Binärsymbolen umfasst,
wobei der zweite Kodierer (6) einen Fehler-zu-Symbol-Wandler (36) zum Empfangen des Voraussagefehlers und Umwandeln des Voraussagefehlers in eine Sequenz von Binärsymbolen, eine zweite Wahrscheinlichkeitsschätzeinheit (26) zum Empfangen der Sequenz von Binärsymbolen und Schätzen einer Wahrscheinlichkeit des Auftretens eines der Binärsymbole und eine zweiten Codewort-Zuordner (16) zum Kodieren der Sequenz von Binärsymbolen umfasst; und
wobei der dritte Kodierer (8) einen zweiten Fehler-zu-Symbol-Wandler (38) zum Empfangen des Voraussagefehlers und Umwandeln des Voraussagefehlers in eine Sequenz von Binärsymbolen, eine dritte Wahrscheinlichkeitsschätzeinheit (28) zum Empfangen der Sequenz von Binärsymbolen und Schätzen einer Wahrscheinlichkeit des Auftretens eines der Binärsymbole und einen dritten Codewort-Zuordner (18) zum Kodieren der Sequenz von Binärsymbolen umfasst. - Kodiervorrichtung (400) nach Anspruch 2, bei der mindestens einer von dem ersten Fehler-zu-Symbol-Wandler (36) und dem zweiten Fehler-zu-Symbol-Wandler (38) aufeinanderfolgend Vergleichswerte erzeugt, beginnend mit einem Wert, der mit größter Wahrscheinlichkeit als ein in einen von dem ersten Fehler-zu-Symbol-Wandler (36) und dem zweiten Fehler-zu-Symbol-Wandler (38) einzugebender Voraussagefehler auftreten würde, aufeinanderfolgend die erzeugten Werte einen nach dem anderen mit dem Voraussagefehler, der in einen von dem ersten Fehler-zu-Symbol-Wandler (36) und dem zweiten Fehler-zu-Symbol-Wandler (38) eingegeben wird, vergleicht und eine Sequenz von Binärsymbolen erzeugt und ausgibt, basierend auf einer Zählung des Vergleichs bis einer der erzeugten Vergleichswerte mit dem Voraussagefehler übereinstimmt.
- Dekodiervorrichtung (500), umfassend:einen Bildelementspeicher (41) zum Speichern von Werten von zu dekodierenden Dekodierbildelementen und Ausgeben von Werten der dekodierten Bildelemente benachbart zu einem Dekodierbildelement als Werte von Referenzbildelementen;eine Modus-Bestimmeinheit (42) zum Bestimmen einer spezifischen Dekodierart oder einer von der spezifischen Dekodierart abweichenden Dekodierart aus einer Vielzahl von vorbestimmten Dekodierarten für das Dekodierbildelement;eine erste Dekodiereinheit (201) zum Empfangen eines Codewortes, Voraussagen eines Wertes des Dekodierbildelementes, Bestimmen eines Voraussageergebnisses, ob die Voraussage korrekt ist oder nicht, und zum Dekodieren des Codewortes in den Wert des Dekodierbildelementes basierend auf dem Voraussageergebnis der Bestimmung bei der bestimmten spezifischen Dekodierart;eine zweite Dekodiereinheit (202) zum Empfangen eines Codewortes für das Dekodierbildelement, Voraussagen des Wertes des Dekodierbildelementes, Dekodieren des Codewortes gemäß der von der spezifischen Dekodierart abweichenden Dekodierart; undeine Dekodier-Steuereinheit (43) zum ausgewählten Betreiben der ersten Dekodiereinheit und der zweiten Dekodiereinheit basierend auf einer von der Modus-Bestimmeinheit (42) bestimmten Dekodierart, undwobei die erste Dekodiereinheit (201) umfasst:eine erste Voraussageeinheit (3) zum Berechnen eines Voraussagewertes des Dekodierbildelementes basierend auf den Werten der Referenzbildelemente;einen ersten Dekodierer (45) zum Dekodieren des Codeworts in eine Binärsymbolsequenz von Voraussageergebnissen der Bestimmung, die angeben, ob die Voraussage korrekt ist oder nicht;einen zweiten Dekodierer (46) zum Dekodieren des das Dekodierbildelement repräsentierenden Codeworts in den Voraussagefehler, wenn die Voraussage nicht korrekt ist;einen ersten Dekodierbildelement-Berechner (32) zum Berechnen des Wertes des Dekodierbildelementes basierend auf dem Voraussagewert des Dekodierbildelementes, der von der ersten Voraussageeinheit (3) berechnet wurde, und dem Voraussagefehler, der von dem zweiten Dekodierer (46) erhalten wurde, undwobei die zweite Dekodiereinheit (202) umfasst:eine zweite Voraussageeinheit (7) zum Berechnen eines Voraussagewertes des Dekodierbildelementes basierend auf den Werten der Referenzbildelemente;einen dritten Dekodierer (48) zum Dekodieren des das Dekodierbildelement repräsentierenden Codewortes in den Voraussagefehler;einen zweiten Dekodierbildelement-Berechner (33) zum Berechnen des Wertes des Dekodierbildelementes basierend auf dem Voraussagewert des von der zweiten Voraussageeinheit (7) berechneten Dekodierbildelementes und dem von dem dritten Dekodierer (48) erhaltenen Voraussagefehler; undwobei die Modus-Bestimmeinheit (42) die Werte der Referenzbildelemente und das Voraussageergebnis der Bestimmung, das von dem ersten Dekodierer (45) dekodiert wurde, empfängt und eine Dekodierart aus der spezifischen Dekodierart, die von der ersten Dekodiereinheit (201) verwendet wird, und der von der spezifischen Dekodierart abweichenden Dekodierart, die von der zweiten Dekodiereinheit (202) verwendet wird, auswählt, derart, dass die spezifische Dekodierart (Mode A) fortgesetzt wird, wenn das Voraussageergebnis der Bestimmung bestimmt, dass die Voraussage sich als korrekt erwiesen hat, wenn das Voraussageergebnis bestimmt, dass die Voraussage sich als nicht korrekt erwiesen hat, das Dekodierbildelement in der spezifischen Dekodierart (Mode A) dekodiert wird und dann die von der spezifischen Dekodierart abweichende Dekodierart (Mode B) festgesetzt wird, die spezifische Dekodierart (Mode A) ausgewählt wird, wenn die Werte der Referenzbildelemente gleich sind, und die von der spezifischen Dekodierart abweichende Dekodierart (Mode B) fortgesetzt wird, wenn die Werte der Referenzbildelemente nicht gleich sind.
- Dekodiervorrichtung (500) nach Anspruch 4, bei der der erste Dekodierer (45) eine erste Symbolwiederherstellungsvorrichtung (55) zum Empfangen des Codewortes und Dekodieren des Codewortes in eine Sequenz von Binärsymbolen und eine erste Wahrscheinlichkeitsschätzeinheit (75) zum Schätzen einer Wahrscheinlichkeit des Auftretens eines der Binärsymbole aufweist und wobei der erste Dekodierer eines der Binärsymbole als Voraussageergebnis der Bestimmung ausgibt,
wobei der zweite Dekodierer (46) eine zweite Symbolwiederherstellungsvorrichtung (56) zum Empfangen des Codewortes und Dekodieren des Codewortes in eine Sequenz von Binärsymbolen, eine zweite Wahrscheinlichkeitsschätzvorrichtung (76) zum Empfangen der Binärsymbole und Schätzen der Wahrscheinlichkeit des Auftretens eines der Binärsymbole und einen ersten Symbol-zu-Fehler-Wandler (86) zum Empfangen der Sequenz von Binärsymbolen und Umwandeln der Sequenz von Binärsymbolen in den Voraussagefehler aufweist, und
wobei der dritte Dekodierer (48) eine dritte Symbolwiederherstellungsvorrichtung (58) zum Empfangen des Codewortes und Dekodieren des Codewortes in eine Sequenz von Binärsymbolen, eine dritte Wahrscheinlichkeitsschätzvorrichtung (78) zum Empfangen der Binärsymbole und Schätzen einer Wahrscheinlichkeit des Auftretens eines der Binärsymbole und einen zweiten Symbol-zu-Fehler-Wandler (88) zum Empfangen der Sequenz von Binärsymbolen und Umwandeln der Sequenz von Binärsymbolen in den Voraussagefehler aufweist. - Dekodiervorrichtung (500) nach Anspruch 5, bei der mindestens einer von dem ersten Symbol-zu-Fehler-Wandler (86) und dem zweiten Symbol-zu-Fehler-Wandler (88) die eingegebene Sequenz von Binärsymbolen in den Voraussagefehler basierend auf einem Wert und einer Zahl von eingegebenen Binärsymbolen umwandelt.
- Kodierverfahren, umfassend:einen Ausgabeschritt zum Empfangen und Speichern eines Wertes eines Kodierbildelementes und zum Ausgeben von Werten von kodierten Bildelementen benachbart zu dem Kodierbildelement als Werte von Referenzbildelementen;einen Modus-Entscheidungsschritt zum Festlegen einer spezifischen Kodierart oder einer anderen Kodierart als die spezifische Kodierart aus einer Vielzahl von vorbestimmten Kodierarten für das Kodierbildelement;einen ersten Hauptkodierschritt zum Voraussagen des Wertes des Kodierbildelementes, Bestimmen eines Voraussageergebnisses, ob die Voraussage korrekt ist oder nicht, Kodieren des Wertes des Kodierbildelementes basierend auf dem Voraussageergebnis der Bestimmung und zum Ausgeben eines Codewortes, das das Kodierbildelement bei der festgelegten spezifischen Kodierart repräsentiert;einen zweiten Hauptkodierschritt zum Voraussagen des Wertes des Kodierbildelementes, Kodieren des Wertes des Kodierbildelementes und zum Ausgeben eines Codewortes, das das Kodierbildelement bei der Kodierart abweichend von der spezifischen Kodierart repräsentiert;einen Kodiersteuerschritt zum ausgewählten Betreiben des ersten Hauptschrittes und des zweiten Hauptschrittes basierend auf einem von dem Bestimmungsschritt bestimmten Kodierart,wobei der erste Hauptkodierschritt umfasst:einen ersten Voraussageschritt zum Berechnen eines Voraussagewertes des Kodierbildelementes basierend auf Werten der Referenzbildelemente;einen ersten Voraussagefehler-Berechnungsschritt zum Berechnen einer Differenz zwischen dem Wert des Kodierbildelementes und dem von dem ersten Voraussageschritt berechneten Voraussagewert als Voraussagefehler;eine Bestimmungsschritt zum Bestimmen, ob der von dem ersten Voraussagefehler-Berechnungsschritt berechnete Voraussagefehler angibt, dass die Voraussage korrekt ist oder nicht, und zum Ausgeben des Voraussageergebnisses der Bestimmung;einen ersten Kodierschritt zum Empfangen und Kodieren mit einer Kodierfolge des Voraussageergebnisses der von dem Bestimmungschritt ausgegebenen Bestimmung und zum Ausgeben eines Codewortes, das die Zahl der eine korrekte Voraussage angebenden Voraussageergebnisse repräsentiert, die vor einem Voraussageergebnis aufgetreten sind, das angibt, dass die Voraussage nicht korrekt ist, oder zum Ausgeben eines Codewortes, wenn die Zahl gleich der Codefolge wird; undeinen zweiten Kodierschritt zum Kodieren des Voraussagefehlers, der von dem ersten Voraussagefehler-Berechnungsschritt berechnet wurde, in das Codewort, wenn die Voraussage nicht korrekt ist, und zum Ausgeben des das Kodierbildelement repräsentierenden Codewortes; undwobei der zweite Kodierschritt umfasst:einen zweiten Voraussageschritt zum Berechnen eines Voraussagewertes des Kodierbildelementes basierend auf Werten der Referenzbildelemente;einen zweiten Voraussagefehler-Berechnungsschritt zum Berechnen einer Differenz zwischen dem Wert des Kodierbildelementes und dem von dem zweiten Voraussageschritt berechneten Voraussagewert als einen Voraussagefehler;einen dritten Kodierschritt zum Kodieren des von dem zweiten Voraussagefehler-Berechnungsschritt berechneten Voraussagefehlers in das Codewort und zum Ausgeben des das Kodierbildelement repräsentierenden Codewortes; undwobei der Modus-Entscheidungsschritt eine Kodierart zwischen der spezifischen Kodierart (Mode A) und der von der spezifischen Kodierart abweichenden Kodierart (Mode B) basierend auf den Werten der Referenzbildelemente und des Voraussageergebnisses des Bestimmungsschrittes auswählt, derart, dass die spezifische Kodierart (Mode A) fortgesetzt wird, wenn der Bestimmungsschritt bestimmt, dass die Voraussage sich als korrekt erwiesen hat, wenn der Bestimmungsschritt bestimmt, dass die Voraussage sich als nicht korrekt erwiesen hat, das Kodierbildelement in der spezifischen Kodierart (Mode A) kodiert wird und dann die Kodierart (Mode B) abweichend von der spezifischen Kodierart festgelegt wird, wobei die spezifische Kodierart (Mode A) gewählt wird, wenn die Werte der Referenzbildelemente gleich sind, und die Kodierart (Mode B) abweichend von der spezifischen Kodierart weitergeführt wird, wenn die Werte der Referenzbildelemente nicht gleich sind.
- Kodierverfahren nach Anspruch 7, bei dem der erste Kodierschritt einen ersten Wahrscheinlichkeitsschätzschritt zum Empfangen des Bestimmungsergebnisses, das von dem Bestimmungsschritt als eine Sequenz von Binärsymbolen ausgegeben wird, und zum Schätzen einer Wahrscheinlichkeit des Auftretens eines der Binärsymbole und einen ersten Codewort-Zuordnungsschritt zum Kodieren der Sequenz von Binärsymbolen umfasst,
wobei der zweite Kodierschritt einen Fehler-zu-Symbol-Wandlungsschritt zum Empfangen des Voraussagefehlers und Umwandeln des Voraussagefehlers in eine Sequenz von Binärsymbolen, einen zweiten Wahrscheinlichkeitsschätzschritt zum Empfangen der Sequenz von Binärsymbolen und Schätzen einer Wahrscheinlichkeit des Auftretens eines der Binärsymbole und einen zweiten Codewort-Zuordnungsschritt zum Kodieren der Sequenz von Binärsymbolen umfasst; und
wobei der dritte Kodierschritt einen zweiten Fehler-zu-Symbol-Wandlungsschritt zum Empfangen des Voraussagefehlers und Umwandeln des Voraussagefehlers in eine Sequenz von Binärsymbolen, einen dritten Wahrscheinlichkeitsschätzschritt zum Empfangen der Sequenz von Binärsymbolen und Schätzen einer Wahrscheinlichkeit des Auftretens eines der Binärsymbole und einen dritten Codewort-Zuordnungsschritt zum Kodieren der Sequenz von Binärsymbolen umfasst. - Dekodierverfahren, umfassend:einen Ausgabeschritt zum Speichern von Werten von zu dekodierenden Dekodierbildelementen und Ausgeben von Werten der dekodierten Bildelemente benachbart zu einem Dekodierbildelement als Werte von Referenzbildelementen;einen Modus-Entscheidungsschritt zum Festlegen einer spezifischen Dekodierart oder einer von der spezifischen Dekodierart abweichenden Dekodierart aus einer Vielzahl von vorbestimmten Dekodierarten für das Dekodierbildelement;einen ersten Hauptdekodierschritt zum Empfangen eines Codewortes, Voraussagen eines Wertes des Dekodierbildelementes, Bestimmen eines Voraussageergebnisses, ob die Voraussage korrekt ist oder nicht, und zum Dekodieren des Codewortes in den Wert des Dekodierbildelementes basierend auf dem Voraussageergebnis der Bestimmung bei der bestimmten spezifischen Dekodierart;einen zweiten Hauptdekodierschritt zum Empfangen eines Codewortes für das Dekodierbildelement und Dekodieren des Codewortes gemäß der von der spezifischen Dekodierart abweichenden Dekodierart; undeinen Dekodier-Steuerschritt zum ausgewählten Betreiben des ersten Hauptdekodierschrittes und des zweiten Hauptdekodierschrittes basierend auf einer bestimmten Dekodierart des Entscheidungsschrittes, undwobei der erste Hauptdekodierschritt umfasst:einen ersten Voraussageschritt zum Berechnen eines Voraussagewertes des Dekodierbildelementes basierend auf den Werten der Referenzbildelemente;einen ersten Dekodierschritt zum Dekodieren des Codeworts in eine binäre Symbolsequenz von Voraussageergebnissen der Bestimmung, die angeben, ob die Voraussage korrekt ist oder nicht;einen zweiten Dekodierschritt zum Dekodieren des das Dekodierbildelement repräsentierenden Codeworts in den Voraussagefehler, wenn die Voraussage nicht korrekt ist;einen ersten Dekodierbildelement-Berechnungsschritt zum Berechnen des Wertes des Dekodierbildelementes basierend auf dem Voraussagewert für das Dekodierbildelement, der von dem ersten Voraussageschritt berechnet wurde, und dem Voraussagefehler, der von dem zweiten Dekodierschritt erhalten wurde, undwobei der zweite Hauptdekodierschritt umfasst:einen zweiten Voraussageschritt zum Berechnen eines Voraussagewertes des Dekodierbildelementes basierend auf den Werten der Referenzbildelemente;einen dritten Dekodierschritt zum Dekodieren des das Dekodierbildelement repräsentierenden Codewortes in einen Voraussagefehler;einen zweiten Dekodierbildelement-Berechnungsschritt zum Berechnen des Wertes des Dekodierbildelementes basierend auf dem von dem zweiten Voraussageschritt berechneten Voraussagewert und dem von dem dritten Dekodierschritt erhaltenen Voraussagefehler; undwobei der Modus-Entscheidungsschritt die Werte der Referenzbildelemente und das Voraussageergebnis der Bestimmung, das von dem ersten Dekodierschritt ausgegeben wurde, empfängt und eine Dekodierart aus der spezifischen Dekodierart, die bei dem ersten Hauptdekodierschritt verwendet wird, und der von der spezifischen Dekodierart abweichenden Dekodierart, die bei dem zweiten Hauptdekodierschritt verwendet wird, auswählt, derart, dass die spezifische Dekodierart (Mode A) fortgesetzt wird, wenn das Voraussageergebnis der Bestimmung bestimmt, dass die Voraussage sich als korrekt erwiesen hat, wenn das Voraussageergebnis bestimmt, dass die Voraussage sich als nicht korrekt erwiesen hat, das Dekodierbildelement in der spezifischen Dekodierart (Mode A) dekodiert wird und dann die von der spezifischen Dekodierart abweichende Dekodierart (Mode B) festgesetzt wird, die spezifische Dekodierart (Mode A) ausgewählt wird, wenn die Werte der Referenzbildelemente gleich sind, und die von der
spezifischen Dekodierart abweichende Dekodierart (Mode B) fortgesetzt wird, wenn die Werte der Referenzbildelemente nicht gleich sind. - Dekodierverfahren nach Anspruch 9, bei dem der erste Dekodierschritt einen ersten Symbolwiederherstellungsschritt zum Empfangen des Codewortes und Dekodieren des Codewortes in eine Sequenz von Binärsymbolen und einen ersten Wahrscheinlichkeitsschätzschritt zum Schätzen einer Wahrscheinlichkeit des Auftretens eines der Binärsymbole und einen Schritt des Ausgebens eines der Binärsymbole als Voraussageergebnis der Bestimmung aufweist,
wobei der zweite Dekodierschritt einen zweiten Symbolwiederherstellungsschritt zum Empfangen des Codewortes und Dekodieren des Codewortes in eine Sequenz von Binärsymbolen, einen zweiten Wahrscheinlichkeitsschätzschritt zum Schätzen einer Wahrscheinlichkeit des Auftretens eines der Binärsymbole und einen ersten Symbol-zu-Fehler-Wandlungsschritt zum Empfangen der Sequenz von Binärsymbolen und Umwandeln der Sequenz von Binärsymbolen in den Voraussagefehler aufweist, und
wobei der dritte Dekodierschritt einen dritten Symbolwiederherstellungsschritt zum Empfangen des Codewortes und Dekodieren des Codewortes in eine Sequenz von binären Symbolen, einen dritten Wahrscheinlichkeitsschätzschritt zum Empfangen der Binärsymbole und Schätzen einer Wahrscheinlichkeit des Auftretens eines der Binärsymbole und einen zweiten Symbol-zu-Fehler-Wandlungsschritt zum Empfangen der Sequenz von Binärsymbolen und Umwandeln der Sequenz von Binärsymbolen in den Voraussagefehler aufweist.
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP01129938A EP1198136A1 (de) | 1996-03-19 | 1997-03-12 | Kodierer,Dekodierer, Kodierverfahren, Dekodierverfahren, und Bildprozessor |
EP00122983A EP1079629A1 (de) | 1996-03-19 | 1997-03-12 | Kodierer,dekodierer, kodierverfahren, dekodierverfahren, und bildprozessor |
EP03006547A EP1320263A1 (de) | 1996-03-19 | 1997-03-12 | Kodierer, Dekodierer, Kodierverfahren, Decodierverfahren, und Bildprozessor |
Applications Claiming Priority (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP6311796 | 1996-03-19 | ||
JP63117/96 | 1996-03-19 | ||
PCT/JP1996/003568 WO1997035422A1 (en) | 1996-03-19 | 1996-12-05 | Encoder, decoder and methods used therefor |
WOPCT/JP96/03568 | 1996-12-05 | ||
PCT/JP1997/000768 WO1997035434A1 (en) | 1996-03-19 | 1997-03-12 | Encoder, decoder, their methods, and image processor |
Related Child Applications (5)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP00122983A Division EP1079629A1 (de) | 1996-03-19 | 1997-03-12 | Kodierer,dekodierer, kodierverfahren, dekodierverfahren, und bildprozessor |
EP03006547A Division EP1320263A1 (de) | 1996-03-19 | 1997-03-12 | Kodierer, Dekodierer, Kodierverfahren, Decodierverfahren, und Bildprozessor |
EP00122983.0 Division-Into | 2000-10-23 | ||
EP01129938.5 Division-Into | 2001-12-17 | ||
EP03006547.8 Division-Into | 2003-03-24 |
Publications (3)
Publication Number | Publication Date |
---|---|
EP0827342A1 EP0827342A1 (de) | 1998-03-04 |
EP0827342A4 EP0827342A4 (de) | 2000-10-04 |
EP0827342B1 true EP0827342B1 (de) | 2013-02-27 |
Family
ID=26404202
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP97907276A Expired - Lifetime EP0827342B1 (de) | 1996-03-19 | 1997-03-12 | Prädiktive Bildkodierung |
Country Status (5)
Country | Link |
---|---|
EP (1) | EP0827342B1 (de) |
JP (1) | JP3228943B2 (de) |
AU (1) | AU697471B2 (de) |
CA (1) | CA2221288C (de) |
WO (1) | WO1997035434A1 (de) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5093349B2 (ja) | 2008-06-27 | 2012-12-12 | 富士通株式会社 | 画像圧縮装置及び画像復元装置 |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4939583A (en) * | 1987-09-07 | 1990-07-03 | Hitachi, Ltd. | Entropy-coding system |
Family Cites Families (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR2379956A1 (fr) | 1977-02-08 | 1978-09-01 | Mitsubishi Electric Corp | Systeme de communications a codage de signaux fac-similes |
JPS5927501A (ja) | 1982-08-04 | 1984-02-14 | シャープ株式会社 | 感湿抵抗素子及びその製造方法 |
JPS6165573A (ja) * | 1984-09-06 | 1986-04-04 | Fujitsu Ltd | 画像信号符号化方式 |
JPS62108663A (ja) * | 1985-11-06 | 1987-05-19 | Fujitsu Ltd | エントロピ−符号化方式 |
JPH01251403A (ja) | 1988-03-31 | 1989-10-06 | Toshiba Corp | 磁気ヘッドおよびこの磁気ヘッドの製造方法 |
JPH0795693B2 (ja) * | 1991-04-25 | 1995-10-11 | 三菱電機株式会社 | 符号化・復号化装置 |
JP3108479B2 (ja) * | 1991-08-28 | 2000-11-13 | 株式会社リコー | 符号化復号化方法およびその装置 |
JP2754970B2 (ja) * | 1991-09-06 | 1998-05-20 | 日本電気株式会社 | 移動体衛星通信システムのデータ同期化方式 |
JP2954438B2 (ja) * | 1992-11-19 | 1999-09-27 | 三菱電機株式会社 | 符号化装置 |
JPH06181523A (ja) * | 1992-12-14 | 1994-06-28 | Ricoh Co Ltd | 予測符号化方式の符号化装置および復号化装置 |
-
1997
- 1997-03-12 WO PCT/JP1997/000768 patent/WO1997035434A1/ja active IP Right Grant
- 1997-03-12 AU AU19395/97A patent/AU697471B2/en not_active Expired
- 1997-03-12 CA CA 2221288 patent/CA2221288C/en not_active Expired - Lifetime
- 1997-03-12 JP JP53334897A patent/JP3228943B2/ja not_active Expired - Lifetime
- 1997-03-12 EP EP97907276A patent/EP0827342B1/de not_active Expired - Lifetime
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4939583A (en) * | 1987-09-07 | 1990-07-03 | Hitachi, Ltd. | Entropy-coding system |
Non-Patent Citations (2)
Title |
---|
BAHL L R ET AL: "IMAGE DATA COMPRESSION BY PREDICTIVE CODING. ÖII: ENCODING ALGORITHMS", IBM JOURNAL OF RESEARCH AND DEVELOPMENT, INTERNATIONAL BUSINESS MACHINES CORPORATION, NEW YORK, NY, US, 1 March 1974 (1974-03-01), pages 172 - 179, XP000577370, ISSN: 0018-8646 * |
KOBAYASHI H ET AL: "IMAGE DATA COMPRESSION BY PREDICTIVE CODING. ÖI: PREDICTION ALGORITHMS", IBM JOURNAL OF RESEARCH AND DEVELOPMENT, INTERNATIONAL BUSINESS MACHINES CORPORATION, NEW YORK, NY, US, 1 March 1974 (1974-03-01), pages 164 - 171, XP000577369, ISSN: 0018-8646 * |
Also Published As
Publication number | Publication date |
---|---|
AU1939597A (en) | 1997-10-10 |
WO1997035434A1 (en) | 1997-09-25 |
EP0827342A4 (de) | 2000-10-04 |
EP0827342A1 (de) | 1998-03-04 |
CA2221288C (en) | 2001-02-27 |
JP3228943B2 (ja) | 2001-11-12 |
AU697471B2 (en) | 1998-10-08 |
CA2221288A1 (en) | 1997-09-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6188793B1 (en) | Encoding apparatus, decoding apparatus, encoding method and decoding method | |
US6954157B2 (en) | Variable length coding method and variable length decoding method | |
US20040008771A1 (en) | Spatial prediction based intra coding | |
WO2003105070A1 (en) | Spatial prediction based intra coding | |
JP2008228329A (ja) | 限定されたビット数を用いる画像処理 | |
US6754393B2 (en) | Encoding apparatus, decoding apparatus, encoding method, and decoding method | |
US6636641B1 (en) | Encoding apparatus, decoding apparatus, encoding method and decoding method | |
EP0827342B1 (de) | Prädiktive Bildkodierung | |
CA2320580C (en) | Encoding apparatus, decoding apparatus, encoding method, decoding method | |
JP3218226B2 (ja) | 符号化装置及び復号装置及びそれらの方法及び画像処理装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
17P | Request for examination filed |
Effective date: 19971117 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): DE DK FR GB IT |
|
A4 | Supplementary search report drawn up and despatched |
Effective date: 20000818 |
|
AK | Designated contracting states |
Kind code of ref document: A4 Designated state(s): DE DK FR GB IT |
|
RIC1 | Information provided on ipc code assigned before grant |
Free format text: 7H 04N 7/24 A, 7H 04N 7/26 B |
|
17Q | First examination report despatched |
Effective date: 20010314 |
|
APBN | Date of receipt of notice of appeal recorded |
Free format text: ORIGINAL CODE: EPIDOSNNOA2E |
|
APBR | Date of receipt of statement of grounds of appeal recorded |
Free format text: ORIGINAL CODE: EPIDOSNNOA3E |
|
APAF | Appeal reference modified |
Free format text: ORIGINAL CODE: EPIDOSCREFNE |
|
APAF | Appeal reference modified |
Free format text: ORIGINAL CODE: EPIDOSCREFNE |
|
RAP1 | Party data changed (applicant data changed or rights of an application transferred) |
Owner name: MITSUBISHI DENKI KABUSHIKI KAISHA |
|
APAF | Appeal reference modified |
Free format text: ORIGINAL CODE: EPIDOSCREFNE |
|
APBT | Appeal procedure closed |
Free format text: ORIGINAL CODE: EPIDOSNNOA9E |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R079 Ref document number: 69740560 Country of ref document: DE Free format text: PREVIOUS MAIN CLASS: H04N0007240000 Ipc: H04N0007340000 |
|
RIC1 | Information provided on ipc code assigned before grant |
Ipc: H04N 7/26 20060101ALI20120716BHEP Ipc: H04N 7/34 20060101AFI20120716BHEP Ipc: H04N 1/41 20060101ALI20120716BHEP |
|
GRAP | Despatch of communication of intention to grant a patent |
Free format text: ORIGINAL CODE: EPIDOSNIGR1 |
|
GRAS | Grant fee paid |
Free format text: ORIGINAL CODE: EPIDOSNIGR3 |
|
GRAA | (expected) grant |
Free format text: ORIGINAL CODE: 0009210 |
|
AK | Designated contracting states |
Kind code of ref document: B1 Designated state(s): DE DK FR GB IT |
|
REG | Reference to a national code |
Ref country code: GB Ref legal event code: FG4D |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R096 Ref document number: 69740560 Country of ref document: DE Effective date: 20130425 |
|
REG | Reference to a national code |
Ref country code: DK Ref legal event code: T3 |
|
PLBE | No opposition filed within time limit |
Free format text: ORIGINAL CODE: 0009261 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT |
|
26N | No opposition filed |
Effective date: 20131128 |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R097 Ref document number: 69740560 Country of ref document: DE Effective date: 20131128 |
|
REG | Reference to a national code |
Ref country code: FR Ref legal event code: PLFP Year of fee payment: 20 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: DE Payment date: 20160308 Year of fee payment: 20 Ref country code: DK Payment date: 20160310 Year of fee payment: 20 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: FR Payment date: 20160208 Year of fee payment: 20 Ref country code: GB Payment date: 20160309 Year of fee payment: 20 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: IT Payment date: 20160324 Year of fee payment: 20 |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R071 Ref document number: 69740560 Country of ref document: DE |
|
REG | Reference to a national code |
Ref country code: DK Ref legal event code: EUP Effective date: 20170312 |
|
REG | Reference to a national code |
Ref country code: GB Ref legal event code: PE20 Expiry date: 20170311 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: GB Free format text: LAPSE BECAUSE OF EXPIRATION OF PROTECTION Effective date: 20170311 |